www.hematologicaloncology.com
Vol. 39 Supp. 1 June 2021
ISSN 0278–0232
Articles from the 16th International Conference on Malignant Lymphoma,
Lugano, Switzerland
Virtual edition, June 1822, 2021
Hematological Oncology


18
Bayer-Sponsored Satellite Symposium at ICML 2021
Friday, 18 June 2021 16:00 – 17:30 CET
16th ICML, Virtual Edition
The Evolving Treatment
Landscape and Emerging
Novel Therapies in
Indolent Non-Hodgkins
Lymphoma (iNHL)
MA-PFM-ONC-ALL-0302-1 03/2021
Bayer (Schweiz) AG, Grubenstrasse 6, 8045 Zurich, Switzerland
Welcome and Opening Remarks
Martin Dreyling, MD (Chair)
University Hospital Großhadern
Ludwig Maximilians-University
Munich, Germany
Current Treatment Landscape and Emerging
Therapies in iNHL
Gilles Salles, MD, PhD
Memorial Sloan Kettering Cancer Center
New York, NY
The Evolving Therapeutic Options for Patients
With Follicular Lymphoma After First Relapse
Pier Luigi Zinzani, MD, PhD
University of Bologna
Bologna, Italy
How to Treat Relapsed Marginal Zone
Lymphoma? Challenges and Clinical
Considerations
Christian Buske, MD
Institute of Experimental Cancer Research
University Hospital Ulm, Ulm, Germany
Live Q&A Session
All panelists
Led by: Martin Dreyling, MD (Chair)
University Hospital Großhadern
Ludwig Maximilians-University
Munich, Germany
Hematological
ONCOLOGY
Editor-in-Chief
Francesco Bertoni


J. Westermann
Berlin, Germany
H. Zhang
Tianjin, China
E. Zucca
Bellinzona, Switzerland
M. Ponzoni
Milan, Italy
D. Rossi
Bellinzona, Switzerland
K. Tobinai
Tokyo, Japan
R.W. Tsang
Toronto, Canada
Govind Bhagat

Hui Wei

Editorial Board
M. Dreyling
Groβhadern, Germany
M. Gramatzki
Kiel, Germany
P. Johnson
Southampton, UK
S. Nakamura
Nagoya, Japan
Andrés J.M. Ferreri

Associate Editors
Disclaimer
Copyright
 e v
nn 
re
evnver
  e
ve-
ve e
n
ee
vn
v
Aims and Scope
Hematological Oncology















01/2021 NP-CH-CAR-0007
Within each and every patient lies the potential
to create a living cell therapy. We are deeply
committed to advancing cell therapy by pursuing
sustainable partnerships in the healthcare
landscape, pioneering a pipeline of innovative
constructs, and utilizing novel manufacturing.
Advancing cell therapy –
for the patients.
WITH EVERY
IN THEIR BEING
PATIENTS GIVE THEIR
ALL. SO DO WE.
Transforming patients’ lives through science
Celgene-CAR-T-Inseratenkampagne-Corporate-A4_EN.indd 1Celgene-CAR-T-Inseratenkampagne-Corporate-A4_EN.indd 1 01.04.21 15:4001.04.21 15:40
Hematological Oncology
Vol. 39 Suppl. 1 June 2021
16th Internaonal Conference on Malignant Lymphoma
Contents
 
 
 
 
MEET THE PROFESSOR ARTICLES
 
 
 

 

 
 
 

 

 

 
EDUCATIONAL SYMPOSIUM ARTICLES
Educaonal symposium on High risk follicular lymphoma
 
 
 
Educaonal symposium on Immunotherapy
 

 
 

Please Visit www.daiichisankyo.com
to Explore Our Oncology Pipeline
Daiichi Sankyo Europe GmbH, Zielstattstr. 48, 81379 Munich, Germany.
© 2021 Daiichi Sankyo Europe GmbH. All Rights Reserved. GLM-GCS-MONC-0049 01/21
DOI: 10.1002/hon.2885
SUPPLEMENT ARTICLE
16
th
International Conference on Malignant Lymphoma
Virtual Edition 18 22 June, 2021
Organized by the Foundation for the Institute of Oncology Research (IOR) in cooperation with the American Association for Cancer Research
(AACR)
Note: The articles published in the 16ICML Educational Book represent the presentations made at the “Meet the Professor” sessions and
“Educational Symposia” during the 16th International Conference on Malignant Lymphoma (16ICML Virtual Edition Lugano, Switzerland) on
June 1822, 2021.
All manuscripts submitted have been subjected to peer review, and authors have been requested to disclose any relationships with the
companies whose products or services are discussed in their manuscripts.
The 16ICML Educational Book production has been supported by an unrestricted educational grant of BeiGene.
Disclaimer: This Educational Book and the Abstract Book have been produced using authorsupplied copy. Editing has been restricted to some
corrections of spelling and style where appropriate. No responsibility is assumed for any claims, instructions, methods or drug dosages contained
in the articles and abstracts: it is recommended that these are verified independently.
Hematological Oncology. 2021;39(S1):714. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
-
7
ACKNOWLEDGEMENTS
INTERNATIONAL CONFERENCE ON MALIGNANT LYMPHOMA (ICML) and FOUNDATION FOR THE INSTITUTE OF ONCOLOGY
RESEARCH (IOR) PRESIDENT
Franco Cavalli, Bellinzona (Switzerland)
ICML ORGANIZING COMMITTEE CHAIRMAN
Francesco Bertoni, Bellinzona (Switzerland)
ICML ORGANIZING COMMITTEE
(based in Bellinzona)
Michele Ghielmini
Alden Moccia
Bertrand Nadel, Marseilles (France) Representative of American Association for Cancer Research (AACR)
Fedro Peccatori, Milan (Italy) Representative of European School of Oncology (ESO)
Davide Robbiani
Davide Rossi
Anastasios Stathis
Georg Stüssi
Emanuele Zucca
16ICML ADVISORY BOARD
James O. Armitage, Omaha, NE (USA) CHAIRMAN
Elias Campo, Barcelona (Spain)
Bruce D. Cheson, Washington, D.C. (USA)
Nicholas Chiorazzi, Manhasset, NY (USA)
Wee Joo Chng, Singapore (Singapore) Riccardo DallaFavera, New York, NY (USA)
Stephan Dirnhofer, Basel (Switzerland)
Martin Dreyling, Munich (Germany)
Jonathan W. Friedberg, Rochester, NY (USA)
Gianluca Gaidano, Novara (Italy)
Philippe Gaulard, Créteil (France)
Mary K. Gospodarowicz, Toronto, ON (Canada)
Peter W.M. Johnson, Southampton (UK)
Brad S. Kahl, St. Louis, MO (USA)Ralf K üppers, Essen (Germany)
John P. Leonard, New York, NY (USA)
T. Andrew Lister, London (UK)
Markus G. Manz, Zurich (Switzerland)
Ari M. Melnick, New York, NY (USA)
Koichi Ohshima, Kurume (Japan)
Astrid Pavlovsky, Buenos Aires (Argentina)
Leticia QuintanillaMartinez de Fend, Tübingen (Germany)
S.Vincent Rajkumar, Rochester, MN (USA)
Kerry J. Savage, Vancouver BC (Canada)
John F. Seymour, Melbourne (Australia)
Margaret A. Shipp, Boston, MA (USA)
Lena Specht, Copenhagen (Denmark)
Catherine Thieblemont, Paris (France)
Pier Luigi Zinzani, Bologna (Italy)
Wilhelm Wössmann, Hamburg (Germany)
Weili Zhao, Shangai (China)
8
-
CONFERENCE SECRETARIAT
Cristiana Brentan, Federica Cariglia, Angela Kurtz, Orietta Lugli, Chiara Saporiti
Foundation for the Institute of Oncology Research (IOR)
Via Vela 6, CH6500 Bellinzona
The Organizers would like to express their sincere gratitude to the Industry Partners who continue to support the Conference. Their unre-
stricted support is essential to the dissemination of the Scientific and Industry programs to the worldwide community engaged in the study and
treatment of lymphoid neoplasms:
AbbVie, Inc.
Acrotech Biopharma LLC
ADC Therapeutics
AstraZeneca
Bayer
BeiGene
Bristol Myers Squibb
Clinigen
CRISPR Therapeutics
Daiichi Sankyo
F. HoffmannLa Roche
Genmab
Gilead Sciences Europe Ltd.
HTG Molecular Diagnostics
Incyte
Janssen Oncology Pharmaceutical Companies of Johnson & Johnson
Karyopharm
Loxo Oncology at Lilly
Lymphoma Hub
Miltenyi Biomedicine
MorphoSys
MSD
Novartis Oncology
Pfizer Oncology Regeneron
Takeda Oncology
(updated on May 2021)
The Organizers would like to express their sincere gratitude to the following Partners for their precious support, essential to the organization of
this Virtual edition of the Conference and to the achievement of its objectives:
American Association for Cancer Research AACR
City of Lugano
European School of Oncology ESO
European School of Oncology Foundation ESOF
European Society for Medical Oncology ESMO
IBSA Foundation for Scientific Research
International Extranodal Lymphoma Study Group IELSG
International Lymphoma Radiation Oncology Group ILROG
Lymphoma Coalition
Saudi Society of Blood & Marrow Transplantation SSBMT
Swiss Cancer Research Foundation
Union for China Lymphoma Investigators UCLI
Università della Svizzera Italiana USI
(updated on May 2021)
-
9
10
-
SCIENTIFIC PROGRAM MEET THE PROFESSOR SESSION AND EDUCATIONAL SYMPOSIUM SCHEDULE
All session times for the 16ICML are Central European Summer Time (CEST)
Saturday, June 19, 2021
Time Channel Session Article/ abstract nr.
13:00 13:30 Channel 1 “Meet the Professor” session
What's new in peripheral Tcell lymphomas EB 06
Stefano Luminari, Reggio Emilia (Italy)
13:30 13:45 Live Q&A
13:00 13:30 Channel 2 “Meet the Professor” session
Cutaneous Tcell lymphoma an update 2021 EB 05
Werner Kempf, Zürich (CH)
13:30 13:45 Live Q&A
13:00 13:30 Channel 3 “Meet the Professor” session
Personalized medicine for Hodgkin lymphoma: mitigating toxicity while preserving cure EB 04
Peter W.M. Johnson, Southampton (UK)
13:30 13:45 Live Q&A
13:00 13:30 Channel 4 “Meet the Professor” session
Use of available prognostic scores in treatment decision: beyond standard prognostic
scores to include molecular/genetics/imaging
John F. Seymour, Melbourne (Australia)
13:30 13:45 Live Q&A
Sunday, June 20, 2021
Time Channel Session Article/ abstract nr.
15:00 16:30 Channel 1 Educational symposium on High risk follicular lymphoma
Cochairs: Elias Campo, Barcelona (Spain) and Carla Casulo, Rochester, NY (USA)
15:00 15:25 Upfront identification of high risk follicular lymphoma EB 12
Carla Casulo, Rochester, NY (USA)
15:25 15:50 Vulnerabilities in the tumor and microenvironment in follicular lymphoma EB 11
Elias Campo, Barcelona (Spain)
15:50 16:15 High risk follicular lymphoma: treatment options EB 10
Brad Kahl, Saint Louis, MO (USA)
16:15 16:30 Live Q&A
16:45 17:15 Channel 1 “Meet the Professor” session
Molecular diagnostics and reporting in lymphoid malignancies: current status and beyond EB 09
Richard Rosenquist Brandell, Stockholm (Sweden)
17:15 17:30 Live Q&A
16:45 17:15 Channel 2 “Meet the Professor” session
New drugs and pharmacological interactions in real life EB 10
Anastasios Stathis, Bellinzona (CH)
17:15 17:30 Live Q&A
16:45 17:15 Channel 3 “Meet the Professor” session
Mantle cell lymphoma advances in molecular biology, prognostication and treatment
approaches
EB 03
Martin Dreyling, Munich (Germany)
(Continues)
-
11
(Continued)
Sunday, June 20, 2021
Time Channel Session Article/ abstract nr.
17:15 17:30 Live Q&A
16:45 17:15 Channel 4 “Meet the Professor” session
Recognizing but not harming. Borderline Bcell lymphoid proliferations EB 07
Leticia QuintanillaMartinez de Fend, Tübingen (Germany)
17:15 17:30 Live Q&A
Monday, June 21, 2021
Time Channel Session Article/ abstract nr.
15:00 16:30 Channel 1 Educational symposium on Immunotherapy
Chair:
15:00 15:25 Optimizing CAR Tcell therapy in lymphoma EB 15
Gilles Salles, New York, NY (USA)
15:25 15:50 Allogeneic stem cell transplant in nonHodgkin lymphomas: still an indication? EB 14
Peter Dreger, Heidelberg (Germany)
15:50 16:15 Bispecific antibodies for the treatment of lymphoma: promises and challenges EB 16
Stephen J. Schuster, Philadelphia, PA (USA)
16:15 16:30 Live Q&A
18:00 18:30 Channel 1 “Meet the Professor” session
Sequencing of myeloma therapy: finding the right path among many standards EB 08
S. Vincent Rajkumar, Rochester, MN (USA)
18:30 18:45 Live Q&A
18:00 18:30 Channel 2 “Meet the Professor” session
Langerhans Cell Histiocytosis: Version 2021 EB 01
Carl E. Allen, Houston, TX (USA)
18:30 18:45 Live Q&A
18:00 18:30 Channel 3 “Meet the Professor” session
Molecular classification of aggressive lymphomas past, present, future EB 02
Björn Chapuy, Göttingen (Germany)
18:30 18:45 Live Q&A
12
-
INDUSTRY PROGRAM SATELLITE SYMPOSIUM SCHEDULE
Friday, June 18, 2021
12:00 13:30 Channel 1 Janssen Oncology Pharmaceutical Companies of Johnson & Johnson
PRECISION MEDICINE IN LYMPHOMA: WHAT DOES IT MEAN?
Chair: John Gribben, London (UK)
12:00 13:30 Channel 2 AstraZeneca
EVOLVING STRATEGIES USING BTK INHIBITORS IN CLL: A SELECTIVE APPROACH TO
IMPROVE PATIENT OUTCOMES
Chair: Paolo Ghia, Milan (Italy)
12:00 13:30 Channel 3 MSD
EXPLORING THE UTILITY OF NOVEL TARGETS IN LYMPHOMAS
Chair: Ulrich Jaeger, Vienna (Austria)
12:00 13:30 Channel 4 ADC Therapeutics
ELEVATE YOUR KNOWLEDGE ON NOVEL ANTIBODYBASED THERAPIES TO TREAT
R/R DLBCL
Chair: Carmelo CarloStella, Milan (Italy)
14:00 15:30 Channel 1 Bristol Myers Squibb
ARE WE GOING TOWARDS A NEW CARE PARADIGM IN FL AND CLL?
Chair: Stefano Luminari, Reggio Emilia (Italy)
14:00 15:30 Channel 2 Medscape (supported by an independent educational grant from Takeda)
SPOTLIGHT ON LYMPHOMA: ADDRESSING REAL WORLD CLINICAL CHALLENGES
FOR HL AND PTCL
Moderator: Timothy Illidge, Manchester (UK)
14:00 15:30 Channel 3 AbbVie
NOT ALL BISPECIFIC ANTIBODIES ARE CREATED EQUAL
Chair: Gilles Salles, New York, NY (USA)
14:00 15:30 Channel 4 Pfizer Oncology
COVID19 IN CLL AND B CELL MALIGNANCIES
Chair: Anthony Mato, New York, NY (USA)
16:00 17:30 Channel 1 Kite, a Gilead Company
BUILDING ON SURVIVAL IN CELL THERAPY: FOCUS ON THE FUTURE
Chair: John Gribben, London (UK)
16:00 17:30 Channel 2 Incyte
PEARLS OF KNOWLEDGE: EXPERT PERSPECTIVES IN THE TREATMENT OF R/R DLBCL
16:00 17:30 Channel 3 Regeneron
EVOLVING TREATMENT PARADIGM FOR RELAPSED/REFRACTORY DIFFUSE LARGE B
CELL LYMPHOMA
Chair: Pier Luigi Zinzani, Bologna (Italy)
16:00 17:30 Channel 4 Bayer
THE EVOLVING TREATMENT LANDSCAPE AND EMERGING NOVEL THERAPIES IN
INDOLENT NONHODGKIN'S LYMPHOMA (iNHL)
Chair: Martin Dreyling, Munich (Germany)
(Continues)
-
13
(Continued)
Friday, June 18, 2021
18:00 19:30 Channel 1 Bristol Myers Squibb
ADVANCEMENTS OF CAR T CELL THERAPIES IN BCELL MALIGNANCIES
Chair: David Maloney, Seattle WA (USA)
18:00 19:30 Channel 2 F. HoffmannLa Roche
UNLOCKING THE FUTURE: SEARCHING FOR A CURE IN DLBCL
Chair: Georg Hess, Mainz (Germany)
18:00 19:30 Channel 3 Lymphoma Hub
SEQUENCING OF THERAPIES IN HIGHRISK RELAPSED/REFRACTORY LYMPHOMA
AND CLL
Chair: Gilles Salles, New York, NY (USA)
18:00 19:30 Channel 2 Novartis
EXPERT PERSPECTIVES IN CART: IDENTIFICATION, REFERRAL AND MANAGEMENT
OF PATIENTS WITH LYMPHOMA
Chair: Anna Sureda, Barcelona (Spain)
14
-
Received: 18 February 2021
DOI: 10.1002/hon.2857
SUPPLEMENT ARTICLE
Langerhans cell histiocytosis: Version 2021
Nitya Gulati
1,2
|Carl E. Allen
1,2
1
Texas Children's Hospital Cancer and
Hematology Center, Houston, Texas, USA
2
Department of Pediatrics, Baylor College of
Medicine, Houston, Texas, USA
Correspondence
Carl E. Allen, Texas Children's Hospital Cancer
and Hematology Center, Houston, TX, USA.
Email: ceallen@txch.org
Abstract
Children with Langerhnans cell histiocytosis (LCH) develop granulomatous lesions
with characteristic clonal CD207+dendritic cells that can arise as single lesions or
lifethreatening disseminated disease. Despite the wide range of clinical pre-
sentations, LCH lesions are histologically indistinguishable based on severity of
disease, and uncertain classification as an immune versus neoplastic disorder has
historically challenged the development of optimal clinical strategies for patients
with LCH. Recently, activating somatic mutations in MAPK pathway genes, most
notably BRAFV600E, have been discovered in almost all cases of LCH. Further, the
stage of myeloid differentiation in which the mutation arises defines the extent of
disease and risk of developing LCHassociated neurodegeneration. MAPK activation
in LCH precursor cells drives myeloid differentiation, inhibits migration, and inhibits
apoptosis, resulting in accumulation of resilient pathologic dendritic cells that re-
cruit and activate T cells. Recurrent somatic mutations in MAPK pathway genes
have also been identified in related histiocytic disorders: juvenile xanthogranuloma,
Erdheim–Chester disease, and Rosai–Dorfman disease. New insights into patho-
genesis support reclassification of these conditions as a myeloid neoplastic disor-
ders. Continued research will uncover opportunities to identify novel targets and
inform personalized therapeutic strategies based on cell of origin, somatic mutation,
inherited risk factors, and residual disease.
KEYWORDS
Erdheim–Chester disease, histiocytic disorder, juvenile xanthogranuloma, Langerhans cell
histiocytosis, Rosai–Dorfman disease
1
|
INTRODUCTION
Histiocytic disorders comprise a heterogeneous group of hematologic
and immunologic conditions historically classified based on histologic
similarities to cells of the mononuclear phagocyte system.
1
However,
with the rapidly increasing understanding of mechanisms of patho-
genesis and ontogeny, a revised classification is proposed that in-
cludes cellular origins, tissue distribution, and molecular lesions along
with histologic features (Tables 1and 2).
2
Langerhans cell histiocy-
tosis (LCH), the most common histiocytic disorder in children, is the
focus of this article. We will also briefly describe featurerelated
disorders: juvenile xanthogranuloma (JXG), Erdheim–Chester
disease (ECD), and sinus histiocytosis with massive lymphadenopathy
(SHML), also called Rosai–Dorfman disease (RDD).
2
|
LANGERHANS CELL HISTIOCYTOSIS
2.1
|
Pathophysiology
LCH has captured the attention of physicians and scientists for
more than 100 years. Clinical cases initially recognized in the early
1900s in children with unusual constellations of bone and pituitary
lesions (Hand–Schüller–Christian disease), aggressive disseminated
Hematological Oncology. 2021;39(S1):1523. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
-
15
disease (Letterer–Siwe disease), or isolated or multifocal bone le-
sions (eosinophilic granuloma). Pathologists in the 1950s noted
histologic similarity of biopsies from patients with these conditions
and proposed a unifying hypothesis that these clinically distinct
syndromes represent a common pathological entity, “Histiocytosis
X”. Subsequently, Nezelof and colleagues identified Birbeck gran-
ules’ pathologic histiocytes of LCH lesions (Figure 1), a feature at
that time was thought to be shared only with epidermal Langer-
hans cells.
LCH lesions are granulomatous lesions consisting of pathologic
“Langerhans cells” (LCs), lymphocytes (primarily Tcells), eosinophils,
and macrophages. Like physiologic epidermal LCs (eLC), LCH lesion
LCs express CD1a and CD207 (langerin) surface markers (Figure 1
and Table 3). Common features among LCH and epidermal LC sup-
ported hypotheses of LCH as a reactive immune disorder, neoplastic
disorder, or some combination of both (reviewed in Allen et al.
4
). In
the 1990s, studies of Xinactivation hinted at the clonal nature of
LCH lesion LCs. In 2010, Rollins and colleagues made a breakthrough
discovery of recurrent somatic BRAFV600E mutations in over 50% of
LCH lesions.
5
Subsequently, alternative BRAF mutations (indels and
fusions) and mutations in MAP2K1 (encoding MEK1) were also been
described. Mutually exclusive somatic activating mutations in MAPK
pathway genes have now been identified in approximately 85% LCH
lesions (Figure 2).
6
As discussed above, shared histology between epidermal LC
and LCH lesion histiocytes prompted updated branding from
“Histiocytosis X” to “Langerhans cell histiocytosis”. However, gene
expression studies comparing LCH lesion CD207+cells to eLC
revealed LCH cells to be relatively less differentiated. Subse-
quently, highsensitivity BRAFV600E PCR assays identified the
mutation in hematopoietic stem cells from bone marrow aspirate
and myeloid precursors from peripheral blood of patients with
disseminated LCH. Notably, BRAFV600E was not identified in
peripheral blood mononuclear cells from patients with single
BRAFV600E+lesions. Enforced expression of BRAFV600E in
langerin +cells in mice resulted in the formation of some
limited LCHlike lesions with minimal impact on overall health,
but enforced expression in CD11c +myeloid cells drove rapid
formation of severe lesions in bone marrow, lung, liver, and
spleen resembling highrisk LCH. Pathologic MAPK activation in
LCH lesion cells results in upregulation of antiapoptotic pro-
gram (BclxL) and downregulation of CCR7, which renders the
cells trapped in lesions, unable to migrate to draining lymph
nodes.
7
Addressing the decadelong debate of LCH pathogenesis
arising from immune dysregulation versus transformation of eLC,
findings over the past decade reframe LCH as a myeloid
neoplastic disorder arising from myeloid precursors.
4
Together,
observations in LCH patients and mouse experiments support a
model of "Misguided Myeloid Differentiation," where state of
differentiation of myeloid precursor in which activating MAPK
mutation arises determines the extent and severity of disease
(Figure 3).
8
TABLE 1Historical classification of histiocytic disorders
Dendritic cell related
Langerhans cell histiocytosis
Juvenile xanthogranuloma/Erdheim–Chester disease
Macrophage related
Hemophagocytic syndromes
Primary hemophagocytic lymphohistiocytosis
Secondary hemophagocytic syndromes
Rosai–Dorfman disease
Malignant diseases
Monocyterelated leukemias
Extramedullary monocytic tumor (myeloid sarcoma)
Macrophagerelated histiocytic sarcoma
Dendritic cell malignancy (malignant histiocytosis)
Note: Adapted from Favara et al.
1
TABLE 2Proposed revised classification of histiocytocytic
disorders
L group LCH
Intermediatecell histiocytosis (ICH)
Erdheim–Chester disease (ECD)
Mixed LCH/ECD
C group Cutaneous nonLCH
Xanthomatous granuloma (XG) family: includes JXG
NonXG family includes cutaneous RDD
Cutaneous nonLCH with major systemic component
XG family: xanthoma disseminatum
NonXG: multicentric reticulohistiocytosis
R group Familial RDD
Sporadic RDD
Classic RDD
Extranodal RDD
RDD with neoplasia or immune disease
Unclassified
M group Primary malignant histiocytosis
Secondary malignant histiocytosis
H group Primary HLH
Secondary HLH (nonMendelian)
HLH of unknown/uncertain origin
Note: Adapted from Emile et al.
2
Abbreviations: ECD, Erdheim–Chester disease; HLH, hemophagocytic
lymphohistiocytosis; JXG, juvenile xanthogranuloma; LCH, Langerhans
cell histiocytosis; RDD, Rosai–Dorfman disease.
16
-
GULATI AND ALLEN
FIGURE 1 Histologic features of Langerhans cell histiocytosis (LCH). Panel A shows typical LCH lesions with large cells, pale cytoplasm,
and reniform nuclei on hematoxylin and eosin staining (A1); CD207positive immunostaining (A2); VE1positive immunostaining for BRAF
V600E protein (A3); and Birbeck granules visualized with electron microscopy (A4). Panel B shows liver involvement, which is frequently
characterized by periportal infiltration by histiocytes (B1) and variable CD207positive staining (B2). Panel C shows biopsy specimens from a
patient with severe LCHassociated neurodegeneration (LCHND), characterized by perivascular VE1positive staining (C1), CD163positive
staining (C2), and a P2RY12 infiltrate with occasional P2RY12positive, tissueresident microglia (C3). Panel D shows histiocytic lesions that
are characteristic of both LCH and juvenile xanthogranuloma (JXG), with heterogeneous histologic features on hematoxylin and eosin staining
(D1), including distinct cell populations that are CD207positive (D2) and CD68positive (D3) (from NEJM, LCH, 379:856–868
3
; Copyright
©2018 Massachusetts Medical Society. Reprinted with permission). JXG, juvenile xanthogranuloma
GULATI AND ALLEN
-
17
2.2
|
Epidemiology
The incidence of LCH is estimated to be approximately 5–10 cases
per million children per year and 1–2 cases per million adults per year
with maletofemale ratio 1.2:1. Registry studies report an increased
incidence in Hispanic populations and rare occurrence of LCH in
children with African ancestry.
9
Notably, a GWAS trio study that
identified an increased risk of LCH in patients with a germline SMAD6
variant, which is enriched in Hispanic populations.
10
2.3
|
Clinical presentation and diagnostic workup
LCH has a wide range of clinical manifestations that can be difficult to
recognize due to overlap with more common conditions (Figure 4).
However, once LCH is considered, diagnosis is fairly straightforward
with biopsy. A stepwise approach to diagnosis and staging is outlined
in Table 4. Characteristic presentations include lytic bone lesions
(80% of cases), rash (20%–40% of cases), soft tissue swelling
(often in proximity to bony lesions), external ear drainage, lymph
node or thymic enlargement, and gum hypertrophy with premature
eruption of teeth. More severe systemic involvement reflected by
cytopenias, heptasplenomegaly, and/or impaired liver function por-
tends a higher risk of morbidity and mortality.Based on the findings
from Histiocyte Society trials, LCH is clinically divided into “high risk”
(liver, spleen, and/or bone marrow involvement) and “low risk” (le-
sions anywhere else), reflecting relative risk of death.
11
The signifi-
cance of “risk sites” in adults remains uncertain.
12
Historical
nomenclature (e.g., Letterer–Siwe) has been replaced with a more
generalized attribution of “LCH” along with description of extent of
disease (e.g., lowrisk single system; lowrisk multisystem; highrisk
multisystem).
Beyond risk attribution, some characteristic sites merit discus-
sion. Skin lesions in infants may represent isolated skin disease with
potential to spontaneously resolve or a component of more extensive
systemic disease requiring chemotherapy. Gastrointestinal tract
involvement is rare but can present with severe diarrhea, hema-
tochezia, malabsorption, and hypoproteinemia. Isolated pulmonary
involvement is more common in young adults with a history of
chronic smoking in the third or fourth decades of life, but is occa-
sionally seen in children and adolescents. Pulmonary involvement
may lead to a severe, chronic debilitating course and often presents
with spontaneous pneumothorax. Central diabetes insipidus (DI) af-
fects approximately 25% of patients, most commonly seen in children
with systemic disease and the orbit and skull involvement. Most
cases of DI present with initial systemic disease, but can also present
as isolated pituitary disease, or arise as a site of relapse. While
posterior pituitary involvement is more common, other endocrine
manifestations associated with LCH may include growth hormone
deficiency, adrenal insufficiency, hyperprolactinemia, or hypogonad-
ism caused by hypothalamic infiltration of the anterior pituitary
gland.
13
LCHassociated neurodegeneration (LCHND) is one of the most
severe complications of LCH. LCHND may develop with the onset of
LCH or several years after the patient has completed therapy and is
presumed to be in remission. The presence of “central nervous sys-
tem (CNS)risk” bone lesions (orbit, mastoid, maxilla, temporal,
sphenoid, zygomatic, and clivus) or pituitary lesions at the time of
initial diagnosis is thought to increase the risk of developing LCHND.
While the true risks of “CNSrisk” require additional investigation,
current practice is generally to treat isolated CNS risk lesions with
systemic chemotherapy. BRAFV600E mutation is also associated with
increased risk of LCHND. Patients with LCHND typically present
with prolonged decline in cognitive abilities, worsening school per-
formance, and/or development of cerebellar symptoms. Character-
istic magnetic resonance imaging (MRI) findings include T2 hyper
intense diffuse or polymorphic lesions involving the white matter of
the cerebellum, pons, basal ganglia, and less often the cerebral
hemispheres (Figure 4).
14
The etiology of LCHND remained uncer-
tain for years, with limited biopsy studies demonstrating lymphocytic
infiltration and activated microglia interpreted as a paraneoplastic or
autoimmune phenomenon. More recent studies demonstrate
microglialike mononuclear cells at sites of neurodegeneration with
BRAFV600E, supporting clonal origin with systemic LCH lesions
(Figure 1).
15
2.4
|
Therapy
2.4.1
|
Local LCH
Treatment options depend on the site and extent of the disease.
Isolated skin lesions sometimes resolve spontaneously, with topical
steroids, or with oral immune suppression (e.g., methotrexate or
hydroxyurea). Single bone lesions in readily accessible and nonCNS
TABLE 3Histologic features of histiocytic disorders
LCH ECD/JXG RDD
HLADR ++ +
CD1a ++
CD14 +/ ++ ++
CD68 +/ ++ ++
CD163 + ++
CD 207 (Langerin) +++
Factor XIIIa ++
Fascin ++ +
Birbeck granules +
Hemophagocytosis +/
Emperiopolesis +
Note: Adapted from Jaffe.
3
Abbreviations: ECD, Erdheim–Chester disease; HLH, hemophagocytic
lymphohistiocytosis; JXG, juvenile xanthogranuloma; LCH, Langerhans
cell histiocytosis; RDD, Rosai–Dorfman disease.
18
-
GULATI AND ALLEN
risks sites may be treated with curettage and/or steroid injection.
Notably, unlike other pediatric “cancer”, LCH does not require
complete excision with “clean” margins—in fact, extensive bone
resection can impair bone remodeling.
2.4.2
|
Frontline therapy
Multiple lesions indicate potential for remote clonal progenitors.
Patients with multiple lesions therefore typically require systemic
chemotherapy. The current standard of care for initial therapy is
vinblastine/prednisone for 1 year (with mercaptopurine added for
highrisk LCH), based on the Histiocyte Society LCHIII trial. LCH
III study demonstrated higher rates of progressionfree survival
(PFS) in patients treated for 1 year versus 6 months (5year PFS
54% vs. 37%; p=0.03) and no benefit of adding methotrexate
for patients with highrisk LCH.
16
Notably, fewer than 50% of
patients with highrisk LCH were cured with vinblastine/predni-
sone. While vinblastine/prednisone may be the current standard,
improved strategies are clearly needed. The Histiocyte Society is
FIGURE 2 MAPK pathway mutations in histiocytic disorders. (A) (left) Schematic of MAPK pathway. Under physiologic conditions, the
growth factor (gray box) engages the tyrosine kinase receptor that transduces the signal to the nucleus. (right) Activating mutations (such as
BRAFV600E) drive constitutive ERK activation. In the case of LCH, this drives the expression of antiapoptosis BCL2L1 (BCLxL) and inhibits
CCR7. (B) Stacked bar graphs represent percentages of MAPK pathway mutations in each histiocytic histologic subtype.
GULATI AND ALLEN
-
19
currently testing the impact of further treatment prolongation (2
vs. 1 year with vinblastine/prednisone/(mercaptopurine) for
frontline therapy (NCT02205762). Another phase 3 trial is
currently randomizing 1 year of vinblastine/prednisone/mercapto-
purine versus 1 year of cytarabine monotherapy for frontline LCH
(NCT02670707).
FIGURE 4 Clinical presentations of Langerhans cell histiocytosis (LCH). Positronemission tomographic (PET) images show a single
bone lesion involving the humerus (Panel A, arrow); lowrisk lesions involving the orbit, lymph nodes, bone (multifocal lesion), and thymus
(Panel B); and highrisk lesions involving the liver, spleen, and bone marrow (Panel C). Other classic presentations include a lytic bone
lesion (Panel D, arrow), cystic lung lesions (Panel E), and various skin lesions (Panels F through I). Examples of LCH lesions involving the
skull and brain include multifocal skull lesions (Panel J, arrow), an orbital lesion (Panel K, arrow), a pituitary lesion (Panel L, arrow), and
LCHassociated neurodegeneration (Panel M, arrow). (from NEJM, LCH, 379:856–8683; Copyright ©2018 Massachusetts Medical
Society. Reprinted with permission)
FIGURE 3 Ontogeny of tissue “histiocytes.” Tissue macrophages and dendritic cells seed tissues at various stages of development.
Microglia, Küpffer cells, and Langerhans cells (LCs) arise from yolksacderived progenitors, followed by fetal liver, then adult hematopoiesis.
Note. Langerhans cells may be generated from adult HSC monocytederived LCs following inflammation or tissue damage. The arrows denotes
the sites of LCH lesions
20
-
GULATI AND ALLEN
2.4.3
|
Salvage therapy
Nucleoside analogs
Optimal approaches for patients who have relapsed or refractory
LCH following frontline therapy have not been established. In two
phase 2 studies, highdose cytarabine/cladribine was effective, but
associated with significant treatmentrelated morbidity and mortal-
ity. By comparison, lowerdose cladribine monotherapy induced high
response rates, but only 3% of patients were cured after 6 months of
therapy. Institutional series support the potential efficacy of
intermediatedose cytarabine or clofarabine monotherapy for
relapsed and refractory LCH. While nucleoside analogs are prom-
ising, prospective trials are needed to determine optimal agent(s),
dose, and duration. Allogeneic hematopoietic cell may also be cura-
tive, but associated with >25% mortality (reviewed in Rodriguez
Galindo and Allen
13
).
Targeted therapy
New concepts of LCH pathogenesis offer an opportunity to move the
treatment for LCH beyond empiricism to rational strategies. Early
adult trials have reported extremely high response rates to MAPK
pathwaydirected targeted therapy with BRAFV600E or MEK inhi-
bition for LCH and ECD.
17,18
Retrospective series report similar
findings for children with LCH. However, despite high response rates,
MAPK pathway inhibition does not appear to be curative.
19,20
Additionally, duration of therapy, durability of response, and poten-
tial for reresponse after stopping the medication, and patterns of
response/resistance are unknown. In LCH patients treated with
MAPK pathway inhibitors, PBMCs with BRAFV600E (presumed
precursors) do not consistently clear from circulation, even in pa-
tients with complete clinical responses, and patient almost univer-
sally relapsed with cessation of therapy.
19,20
Thus, there is an urgent
need to improve therapeutic strategies (e.g., MAPK inhibitors with
TABLE 4Clinical evaluations for newly diagnosed LCH
Initial evaluation
History, physical examination
Laboratory studies
a. Complete blood count
b. Serum chemistry
c. Liver function test
d. Sedimentation rate (ESR)
e. Lactate dehydrogenase (LDH)
f. Serum ferritin
g. Immunoglobulin profile
h. PT/INR, aPTT (with evidence of liver dysfunction)
Imaging
a. Skeletal survey with two views of chest and four views of the skull (if PET/CT not obtained).
Confirmation of LCH and extent of disease evaluation
PET/CT
Diagnostic biopsyexcisional biopsy is preferred. Curettage of bone lesions is optimal and complete excision is not required. The presence of
abnormal clusters of CD1a+/CD207+histiocytes are diagnostic (Note: normal skin and lymph node biopsies may include scattered physiologic CD207+
Langerhans cells). CD163, fascin, and factor XII help identify mixed histiocytic lesions (such as JXG/LCH, ECD/LCH)
Additional studies(based on lab and/or clinical features)
All patients <2 years of age, any patients with cytopenias, liver and spleen involvement
a. Bilateral bone marrow aspirate and biopsy
Any skull lesions (based on physical exam or lytic lesions on skull xrays)—CT skull/maxillofacial scans
CNSrisk lesions
a. MRI brain with and without contrast for patients with CNSrisk lesions
b. For auditory canal or temporal bone involvement, also perform a hearing evaluation
c. For clinical suspicion of DI, pituitary dysfunction, or thickened pituitary stalk on MRI brain–urine specific gravity, urine and serum osmolality/
water deprivation test. (Note: If there is an isolated DI and thickened pituitary stalk without any other features suggestive of LCH, perform diagnostic LP
for cytology and AFP/BHCG to rule out germ cell tumor).
d. Other endocrine evaluation as indicated
e. Baseline neurocognitive evaluation for patients with DI or evidence of LCHND
Spinal cord or vertebral involvementMRI spine with and without contrast
Pulmonary involvementCT chest (chest xrays and PET/CT may miss small pulmonary nodules, cysts, or thymic involvement).
Elevated transaminases, elevated direct bilirubin or decreased albuminabdominal US or MRI
History of malabsorption or hypoalbuminemialower GI endoscopy
Abbreviations: DI, diabetes insipidus; ECD, Erdheim–Chester disease; HLH, hemophagocytic lymphohistiocytosis; JXG, juvenile xanthogranuloma; LCH,
Langerhans cell histiocytosis; LCHNH, LCHassociated neurodegeneration; RDD, Rosai–Dorfman disease.
GULATI AND ALLEN
-
21
chemotherapy or antiapoptotic agents or epigenetic modifiers) to
safely cure patients with LCH and related disorders.
2.5
|
Adult LCH
The natural history of LCH in adults is understudied. Except for the
predominance of lung disease, LCH appears to involve the same
potential organ distribution as seen in children, though the incidence
may be different. For example, pulmonary LCH usually occurs as a
singlesystem disease in patients, 90% of the cases in adults who are
heavy chronic smokers. Other differences include a higher incidence
of oral and genital mucosa involvement in adults. Whole exome
sequencing studies demonstrate higher somatic mutation burden in
adult versus pediatric LCH (where median exome mutation is 1).
When LCH arises de novo in adults, it may reflect acquisition of
mutation(s) through clonal hematopoiesis, reflected by mixed
phenotype myeloproliferative neoplastic disorders in some patients.
For adults with single LCH lesions, management strategies
similar to the pediatric population include curettage (clean margins
are not required) with or without intralesional corticosteroids for
singlebone lesions. However, there is no standard of care for the
management of multisystem disease. Vinblastine/prednisone may
have higher toxicity in adult patients, favoring alternatives such as
cytarabine monotherapy. Other options include cladribine, clofar-
abine, hydroxyurea, methotrexate, 6MP, and MAPK pathway
inhibitors.
12
2.6
|
NonLangerhans cell histiocytic disorders
Juvenile xanthogranuloma (JXG) is a histiocytic disorder that
shares many features with macrophage histology, and is histo-
logically indistinguishable from ECD (Table 3). It most commonly
affects infants and young children with a slight male predomi-
nance and presents as one or more “fleshy skin nodules.”
However, in some patients, it may be systemic (<5% pediatric
cases), involving multiple organs including deeper soft tissue,
CNS, bone, lung, liver, spleen, pancreas, adrenal glands, in-
testines, kidneys, lymph nodes, bone marrow, orbit, and heart.
The etiology and basis for prevalence in children are not well
defined. An association between JXG and neurofibromatosis
(types 1 and 2) and juvenile myelomonocytic leukemia and other
molecular alterations such as CSF1R,KRAS,NRAS, and MAP2K1
(rarely BRAFV600E) implicates constitutive MAPK activation in
pathogenesis (Figure 2).
6
Erdheim–Chester disease (ECD) typically arises in patients between
the ages of 40 and 70, with a male predominance. Typical pre-
sentations include xanthelasma on the upper eyelids, skin rash, and
bilateral lower limb bone pain. Consensus diagnostic criteria for ECD
require the presence of (1) foamy CD68+/CD1ahistiocytes
(Table 3), often with admixed inflammation and fibrosis, and
(2) radiographic findings of bilateral and symmetric abnormalities in
the diaphyseal and metaphyseal regions of the long bones of the legs.
More severe manifestations can include cardiopulmonary insuffi-
ciency, renal failure due to retroperitoneal/perinephric infiltration, or
CNS symptoms such as cerebellar signs, DI, or cognitive dysfunction.
Like LCH, ECD is characterized by activating somatic MAPK pathway
gene mutations (BRAFV600E being the most common in approxi-
mately 50% of patients) (Figure 2). ECD historically carried a dismal
prognosis that has been significantly improved with MAPK inhibitors.
In 2017 vemurafenib was approved for BRAFV600E +ECD; and
MEK inhibition has also been reported with high response rates
(reviewed in Goyal et al.
21
).
RDD or SHML likely represents a variety of conditions that share
pathologic CD68+/CD1acells with emperipolesis (trafficking of
viable lymphocytes through histiocytes) (Table 3). It can occur either
as an isolated disorder or in conjunction with another autoimmune,
malignant, or hereditary disease. Chronic, painless, massive cervical
lymphadenopathy is the most common presentation. Other nodal
areas and extranodal sites (such as skin, upper respiratory mucosa,
ocular structures, bones, and CNS) may also be involved. Treatment
is variable and depends on the number of involved sites and can
range from observation to systemic chemotherapy. MAPK mutations
have been reported in RDD lesions, but less reliably than the other
histiocytic disorders discussed above (Figure 2) (reviewed in Abla
et al.
22
).
3
|
CONCLUSIONS
Clinical advances for patients with LCH and related disorders
have historically been stalled by undefined mechanisms of path-
ogenesis. However, accelerated advances over the past decade
have defined LCH as an inflammatory myeloid neoplastic disorder
with extent of disease determined by the cell of origin in which
activating MAPK somatic mutations arise. The challenge we now
face is how to translate biological discovery into improved out-
comes for children and adults with histiocytic disorders.
Continued research on LCH, JXG, ECD, and RDD will uncover
opportunities to identify novel targets and inform personalized
therapeutic strategies based on cell of origin, somatic mutation,
and inherited risk factors.
ACKNOWLEDGMENT
Dr. Allen receives support from the HistioCure Foundation, St. Bal-
drick's Foundation (Innovation Grant, Consortium Grant), the Moves
for Miles Childhood Cancer Foundation, and the Leukemia and
Lymphoma Society (Translational Research Program).
CONFLICT OF INTEREST
The authors have no conflicts to report with respect to this
manuscript.
ORCID
Carl E. Allen https://orcid.org/0000-0002-6625-739X
22
-
GULATI AND ALLEN
REFERENCES
1. Favara BE, Feller AC, Pauli M, et al. Contemporary classification of
histiocytic disorders. Med Pediatr Oncol. 1997;29:157166.
2. Emile JF, Abla O, Fraitag S, et al. Revised classification of histiocy-
toses and neoplasms of the macrophagedendritic cell lineages.
Blood. 2016;27:26722681.
3. Jaffe R. The diagnostic histopathology of Langerhans cell histiocy-
tosis. In: Weitzman S, Egeler RM, eds. Histiocytic Disorders of Children
and Adults. Basic Science Clinical Features, and Therapy. Cambridge
University Press; 2005:1439.
4. Allen CE, Merad M, McClain KL. Langerhanscell histiocytosis. N Engl
J Med. 2018;379:856868.
5. BadalianVery G, Vergilio JA, Degar BA, et al. Recurrent BRAF
mutations in Langerhans cell histiocytosis. Blood. 2010;116:
19191923.
6. Durham BH, Lopez Rodrigo E, Picarsic J, et al. Activating mutations
in CSF1R and additional receptor tyrosine kinases in histiocytic
neoplasms. Nat Med. 2019;25:18391842.
7. Hogstad B, Berres ML, Chakraborty R, et al. RAF/MEK/extracellular
signalrelated kinase pathway suppresses dendritic cell migration
and traps dendritic cells in Langerhans cell histiocytosis lesions. J Exp
Med. 2018;215:319336.
8. Berres ML, Lim KPH, Peters T, et al. BRAFV600E expression in
precursor versus differentiated dendritic cells defines clinically
distinct LCH risk groups. J Exp Med. 2014;211:669683.
9. Ribeiro KB, Degar B, Antoneli CBG, Rollins B, RodriguezGalindo C.
Ethnicity, race, and socioeconomic status influence incidence of
Langerhans cell histiocytosis. Pediatr Blood Cancer. 2015;62:
982987.
10. PeckhamGregory EC, Chakraborty R, Scheurer ME, et al. A
genomewide association study of LCH identifies a variant in
SMAD6 associated with susceptibility. Blood. 2017;130:22292232.
11. Gadner H, Grois N, Pötschger U, et al. Improved outcome in multi-
system Langerhans cell histiocytosis is associated with therapy
intensification. Blood. 2008;111:25562562.
12. Girschikofsky M, Arico M, Castillo D, et al. Management of adult
patients with Langerhans cell histiocytosis: recommendations from
an expert panel on behalf of EuroHistioNet. Orphanet J Rare Dis.
2013;8:72.
13. RodriguezGalindo C, Allen CE. Langerhans cell histiocytosis. Blood.
2020;135:13191331.
14. Yeh EA, Greenberg J, Abla O, et al. Evaluation and treatment of
Langerhans cell histiocytosis patients with central nervous system
abnormalities: current views and new vistas. Pediatr Blood Cancer.
2017;65.
15. McClain KL, Picarsic J, Chakraborty R, et al. CNS Langerhans cell
histiocytosis: common hematopoietic origin for LCHassociated
neurodegeneration and mass lesions. Cancer. 2018;124.
16. Gadner H, Minkov M, Grois N, et al. Therapy prolongation improves
outcome in multisystem Langerhans cell histiocytosis. Blood.
2013;121.
17. Diamond EL, Subbiah V, Lockhart AC, et al. Vemurafenib for BRAF
V600Mutant ErdheimChester disease and Langerhans cell histio-
cytosis: analysis of data from the histologyindependent, phase 2,
Openlabel VEBASKET study. JAMA Oncol 2017;4(3).
18. Haroche J, CohenAubart F, Emile JF, et al. Dramatic efficacy of
vemurafenib in both multisystemic and refractory ErdheimChester
disease and Langerhans cell histiocytosis harboring the BRAF V600E
mutation. Blood. 2013;121:14951500.
19. Donadieu J, Larabi IA, Tardieu M, et al. Vemurafenib for refractory
multisystem Langerhans cell histiocytosis in children: an interna-
tional observational study. J Clin Oncol. 2019;JCO1900456.
20. Eckstein OS, Visser J, RodriguezGalindo C, Allen CE. Clinical re-
sponses and persistent BRAFV600E+blood cells in children with
LCH treated with MAPK pathway inhibition. Blood. 2019;133(15):
16911694.
21. Goyal G, Heaney ML, Collin M, et al. ErdheimChester disease:
consensus recommendations for evaluation, diagnosis, and treat-
ment in the molecular era. Blood. 2020;135:19291945.
22. Abla O, Jacobsen E, Picarsic J, et al. Consensus recommendations for
the diagnosis and clinical management of RosaiDorfman
Destombes disease. Blood. 2018;131:28772890.
How to cite this article: Gulati N, Allen CE. Langerhans cell
histiocytosis: Version 2021. Hematological Oncology.
2021;39(S1):1523. https://doi.org/10.1002/hon.2857
GULATI AND ALLEN
-
23
Received: 17 February 2021
DOI: 10.1002/hon.2847
SUPPLEMENT ARTICLE
Molecular classification of aggressive lymphomas—past,
present, future
Kirsty Wienand |Björn Chapuy
Department of Hematology and Medical
Oncology, University Medical Center
Göttingen, Göttingen, Germany
Correspondence
Björn Chapuy, Department of Hematology
and Medical Oncology,University Medical
Center Göttingen RobertKoch Str. 40, 37075
Göttingen, Germany.
Email: bjoern.chapuy@med.unigoettingen.de
Abstract
Aggressive large Bcell lymphomas (LBCLs) represent a frequent but clinically and
molecularly heterogeneous group of tumors. Technological advances over the last
decades prompted the development of different classification schemas to either
sharpen diagnoses, dissect molecular heterogeneity, predict outcome, or identify
rational treatment targets. Despite increased diagnostic precision and a noticeably
improved molecular understanding of these lymphomas, clinical perspectives of pa-
tients largely remain unchanged. Recently, finished comprehensive genomic studies
discovered genetically defined LBCL subtypes that predict outcome, provide insight
into lymphomagenesis, and suggest rational therapies with the hope of generating
patienttailored treatments with increased perspective for patients in greatest need.
Current and future efforts integrate multiomics studies and/or leverage singlecell
technologies and will provide us with an even more finegrained picture of LBCL
biology. Here, we highlight examples of how highthroughput technologies aided in a
better molecular understanding of LBCLs and provide examples of how to select
rationally designed targeted treatment approaches that might personalize LBCL
treatment and eventually improve patients' perspective in the near future.
KEYWORDS
biomarker, diffuse large Bcell lymphoma, highgrade Bcell lymphoma, immune escape,
molecular classification, precision medicine, primary central nervous system, primary
mediastinal large Bcell lymphoma, primary testicular lymphoma
1
|
CAPTURING MOLECULAR HETEROGENEITY
OF AGGRESSIVE BCELL LYMPHOMA
Large Bcell lymphomas (LBCLs) represents a clinically and molecu-
larly heterogeneous group of aggressive nonHodgkin lymphomas
that largely arise from antigen exposed B cells and include primary
mediastinal large Bcell lymphoma (PMBL), primary central nervous
system lymphoma (PCNSL), primary testicular lymphoma (PTL), and
diffuse large Bcell lymphoma (DLBCL).
1
Of those, DLBCL often in-
volves multiple nodal and extranodal sites, while PMBL, PCNSL, and
PTL present as localized masses in extranodal sites (Figure 1). Over
the last decade, the molecular heterogeneity of these tumors was
captured in various classification schemes that (i) improved the ac-
curacy of diagnosis; (ii) identified relevant molecular subtypes; (iii)
allowed the development of prognostic and/or predictive markers;
and (iv) provided insight into the associated biology.
Technological advances in recent years have the allowed
detection and prioritization of all genetic alterations types—
recurrent mutations, somatic copy number alterations (SCNAs) and
structural variants (SVs)—followed by integration and assessment of
This is an open access article under the terms of the Creative Commons AttributionNonCommercialNoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is noncommercial and no modifications or adaptations are made.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
24
-
Hematological Oncology. 2021;39(S1):2430. wileyonlinelibrary.com/journal/hon
their temporal ordering and associated transcriptional profiles.
These comprehensive genomic profiles at basepair resolution of
hundreds of patients diagnosed with primary LBCL has captured the
complex intrinsic genetic heterogeneity of these lymphomas, high-
lighted similarities and differences across entities, identified previ-
ously not defined molecular subtypes, provided better insight into
the underlying biology, led to the discovery of rationaltargeted
therapies and might be used to stratify patients for treatments
(Figure 1).
2
|
PRIMARY MEDIASTINAL LARGE BCELL
LYMPHOMAS
PMBL is a rare LBCL that predominantly occurs in young women and
was initially categorized as a morphological subtype of DLBCL.
1
Transcriptional profiling of PMBL revealed reduced expression of
Bcell receptor (BCR) signaling components, including surface
immunoglobulin, and constitutive activation of Janus kinase (JAK)/
signal transducer and activator of transcription (STAT) and nuclear
factor‐κB (NF‐κB) signaling cascades, molecular features reminiscent
of classical Hodgkin lymphoma (cHL) (Figure 1).
2–4
The shared clin-
ical, pathomorphological, and molecular features of PMBL and cHL,
and differences between PMBL and DLBCL led to the recognition of
PMBL as a distinct lymphoma entity.
1
Subsequent studies have defined genetic mechanisms underly-
ing the activation of NF‐κB and JAK/STAT signaling pathways,
including the identification of the recurrently amplified region on
9p/9p24.1, which includes JAK2 and increases JAK/STAT signaling
in PMBL.
5–8
Interestingly, the 9p24.1 amplified region also affected
the PD1 ligands, namely PDL1 (CD274) and PDL2 (PDCD1LG2),
providing a genetic mechanism of immune escape in PMBL.
4,5,8
Notably, the coamplification of PDL1/PDL2 and JAK2 increases the
expression of the PD1 ligands both directly, by increasing their
copy number, and indirectly, increases the ligands' expression via
JAK/STAT signaling and thereby result in an effective immune
escape.
5,9
SVs deregulating PDL1 and PDL2 by translocation,
inversion, or deletion of the 30untranslated region of PDL1 have
also been reported in PMBL.
10
The success of immune checkpoint
blockade with PD1 blocking agents in cHL
11–14
and these striking
genetic similarities between PMBL and cHL prompted the clinical
evaluation of PD1 blockade in PMBL. The high response rate of
single PD1 blockade with pembrolizumab in relapsed/refractory
PMBL led to rapid FDA approval in 2018 and underscored the
importance of characterizing targetable genetic vulnerabilities in
this disease and related lymphoma.
15,16
Interestingly not all patients
with genetic bases of PD1 deregulation responded to PD1
blockade
15,16
which resulted in additional clinical trials that com-
bined PD1 blockade with other rational targets, such as the anti
CD30 antibody–drug conjugate, brentuximab vedotin, with prom-
ising responses and duration.
15
The clinical heterogeneity in responses to PD1 blockade of
PMBL warranted further molecular understanding and resulted in
comprehensive genomic studies that captured somatic mutations,
SCNAs and SVs and prioritized candidate cancer genes (CCGs).
17,18
These studies highlighted that CCGs can be perturbed by more
than one genetic mechanism and that multiple genetic alterations
converge on specific signaling pathways, including the NF‐κB and
JAK/STAT signaling pathways (Figure 1).
17,18
One of these studies
highlighted previously unappreciated aspects of the PMBL genetic
signature that was shared with those in cHL, and could be asso-
ciated with an increased response rate to PD1blockade, including
frequent genetic alterations affecting the ligands of the PD1 (PD
L1/PDL2) with increased PD1 expression, and a relatively high
mutational burden, compared to that in other lymphoid and solid
cancers.
17,19
Additionally, both PMBLs and cHLs exhibit microsat-
ellite instability and APOBEC mutational signatures that have been
associated with a more favorable response to PD1 blockade.
20,21
On the other side, PMBLs, like cHLs, exhibited multiple genetic
bases of perturbed major histocompatibility complex (MHC) class I
expression and less frequently decreased MHC class II expression
that also impacted the cellular composition of the tumor micro-
environment
17
and likely could be associated with reduced
response to PD1 blockade. Importantly, these studies provided a
genomic framework to comprehensively assess molecular mecha-
nisms of response and resistance to PD1 blockade in subsequent
studies and provide the basis for additional rational combination
therapies.
FIGURE 1 Aggressive lymphoma is classified into different entities and subtypes based on morphology/clinically, transcriptomic,
cytogenetic, and genomic features
WIENAND AND CHAPUY
-
25
3
|
PRIMARY CENTRAL NERVOUS SYSTEM
LYMPHOMA AND PRIMARY TESTICULAR
LYMPHOMA
PCNSL and PTL, that both arise in sites previously described as im-
mune sanctuaries, are primary extranodal LBCLs with inferior re-
sponses to current empiric treatment regimens.
22,23
Prior studies had
focused on individual epigenetic or genetic features and transcrip-
tional profiling had suggested that PCNSL/PTL share similarities to
the activated Bcell (ABC) subtype of systemic DLBCL (Figure 1).
24
To gain insight into potential actionable genetic alterations a
comprehensive study characterized the genomic landscape of PCNSL
and PTL and compared the derived genetic alterations to those in
DLBCL and PMBL. That study identified unique combinations of ge-
netic alterations (i.e., genetic signatures) for discrete LBCL subtypes
and putative bases for targeted therapy.
8,25
For instance, like PMBL,
PCNSL, and PTL exhibited frequent 9p24.1/PDL1/PDL2 genetic
alteration,
8
providing a genetic basis of affecting the immune
checkpoint pathway (Figure 1). Encouraged by the high response
rates in lymphomas with genetic bases of PD1 deregulation (cHL and
PMBL), checkpoint inhibition with PD1 blocking antibodies is
currently being tested clinically.
26
PCNSL and PTL, like DLBCL,
frequently exhibited genomic instability, but the molecular mecha-
nisms operating in PTL and PCNSL are distinct from DLBCL and
largely caused by highlevel copy losses of the tumorsuppressor
CDKN2A,
8
while in DLBCL several monoallelic alterations in p53
modifiers were observed.
27
Additionally, PCNSL and PTL harbored
frequent MYD88
L265P
mutations and NFKBIZ copy gains leading to
nearuniform oncogenic Tolllike receptor (TLR) activation.
8
Notably,
most of the PCNSLs had mutations in the proximal BCRassociated
gene, CD79B
8,25
similarly to a subset of ABC–DLBCLs (Figure 1).
28–30
This mutational pattern is reminiscent of a discrete molecular sub-
type of DLBCL that is depending on oncogenic TLR signaling
involving an endosomal multiprotein super complex (i.e., MyTBCR
complex).
29–31
The importance of this oncogenic pathway in PCNSL is
further substantiated by recent insights that highlighted the role of
autoantigens and associated chronic BCR stimulation for the brain
tropism of malignant PCNSL B cells.
32,33
Consistently, the clinical
evaluation of targeted inhibitors against the BCRand TLRsignaling
pathways, including BTK inhibitors, such as ibrutinib alone or in
conjunction with intensive chemoimmunotherapy,
34
or lenalidomide
and derivatives are under clinical evaluation.
35
4
|
HIGH GRADE BCELL LYMPHOMA WITH MYC
TRANSLOCATION AND/OR BCL2 TRANSLOCATION
AND/OR BCL6 TRANSLOCATION
In 2016, the World Health Organization (WHO) has recognized a
rare but highgrade Bcell lymphoma with MYC and BCL2 and/or
BCL6 rearrangements to be included in a single category (high grade
Bcell lymphoma), also known under the prior clinical acronym
“double hit”/ “triple hit” lymphoma, and suggested a provisional entity
(Figure 1).
1,36
According to WHO guidelines
1,36
, all aggressive Bcell
tumors with a germinal center Bcell (GCB) phenotype or highgrade
morphological features or to cases with more than 40% MYC
expressing cells should be tested by Fluorescence in situ hybridiza-
tion (FISH) for translocations in MYC,BCL2, and BCL6 and in case of
double positivity grouped in this categories. Due to different tech-
niques, cut offs and less standardization these tumors are less well
defined. Currently, this is an active area of research in which several
groups attempt to define standards and capture some of the biology
by geneexpressionbased or genetic platforms. However, while
initial reports suggested an dismal clinical outcome,
37–39
more recent
reports suggests that with more frequent testing the outcome be-
comes less dismal.
29,40,41
A recent multiinstitutional intranational
retrospective study reanalyzed all patients with this genotype using
FISH and highlighted that the negative prognostic impact is only
present when MYC was juxtaposed to the IgH enhancer.
41
This data
suggests that in the initial reports were probably based on some
selection bias and that a comprehensive characterization of these
lymphomas is still needed as it is likely that this category itself re-
flects a heterogenous group of tumors.
5
|
DIFFUSE LARGE BCELL LYMPHOMA
DLBCL is the most common aggressive LBCL accounting for up to
35% of all lymph node cancers. Although the majority of patients with
DLBCL are curable with combination immunochemotherapy con-
sisting of rituximab, cyclophosphamide, doxorubicin, vincristine, and
prednisolone (RCHOP), a substantial fraction of patients develop
recurrent or progressive disease that is often fatal. This clinical
heterogeneity had inspired decades of researchers to develop clinical
and/or molecular classifiers capturing the biology, of which several
provide some prognostic guidance and others help to dissect the
molecular underpinnings of the disease. Some of the new next gen-
eration sequencing (NGS)based classifications might even pave the
way to become the first molecular predictive markers in this disease.
5.1
|
Clinical heterogeneity
Clinically, the most widely used risk model is the international
prognostic index (IPI), that has been established in the last century,
42
and with minimal changes, that is until today the most robust clinical
model used in daily routine.
43
Key criticism of the IPI is the missing
link to molecular features that might be amenable to targeted
therapy.
5.1.1
|
Morphological heterogeneity
Initial attempts to leverage morphological subtypes and explore their
power for classification is not extensively used with the exception of
recognizing Tcell/histiocyterich large Bcell lymphoma (THRBCL).
26
-
WIENAND AND CHAPUY
THRBCL is a morphological subtype, in which the rare malignant
DLBCL cell is surrounded by a brisk but ineffective inflammatory
background (Figure 1).
1
For that reason, earlier studies explored
the genetic similarities between TCRLBCL and nodular lymphocyte
predominant Hodgkin lymphoma.
44
Recent works further support a
genetic relationship to cHL and PMBL, as TCRLBCL also exhibits
frequent PD1 ligand expression and a genetic bases of PD1medi-
ated immune escape.
45,46
5.1.2
|
Transcriptional heterogeneity
Besides recognized morphological subtypes,
1
the heterogeneity of
DLBCL is also partially captured in transcriptionally defined sub-
types that provide insights into disease pathogenesis and candidate
treatment targets.
28,47–51
The celloforigin (COO) classification
identifies ABCand GCBtype DLBCLs,
47,52
based on whole tran-
scriptome gene expression profiling and similarities to normal B
cell counterparts (Figure 1).
53
An alternative transcriptional clas-
sification subdivides primary DLBCLs into BCR signaling, oxidative
phosphorylation, and host response DLBCLs based on unsuper-
vised consensus clustering solely in the space of DLBCLs without
comparison to normal tissue (consensus cluster classification;
Figure 1).
48
Also this transcriptional classification prompted in
subsequent studies the discovery of BCR and metabolic de-
pendencies of the respective DLBCL subsets.
50,51,54
The widely
used ABC and GCB distinction had also provided insights into the
biological differences with increase baseline NF‐κB signaling and
an differentiation block towards plasma cells in ABCtype DLBCLs
and an epigenetic locking into the germinal center program in GCB
DLBCLs (for detailed biology reviews on ABC/GCBs see Refer-
ences 52 and 56). In particular, driven by outcome differences
between the two transcriptional phenotypes in some studies (ABC,
inferior clinical outcome), a plethora of parsimonious assays
including immunohistochemistry (IHC),
55
digital gene expression
such as nanostringbased assays
56
or NGS approaches have been
developed. Despite numerous issues regarding accuracy and
reproducibility of IHCbased assays, it is extensively used in
practice and biomarkerdriven studies. Thus far, patient stratifica-
tion for treatment based on transcriptional subtypes either by IHC
or gene expressionbased stratifications have been unsuccessful.
Moreover, although patients with ABC DLBCLs are reported to
have less favorable responses to standard therapy than those with
GCB DLBCLs,
40,47,57
targeted analyses of select alterations have
indicated that there is additional undefined genetic
complexity
27,49,58–60
and thus prompted further comprehensive
genetic studies.
5.1.3
|
Genetic heterogeneity of DLBCL
Initial genetic studies on DLBCL were focused on the discovery of
single genetic alterations, often in the context of COOdefined
transcriptional subtypes.
1,27,61–65
Advances in genomewide tech-
nologies and integrative genomic analysis has allowed the charac-
terization of a more precise picture of the DLBCL genome, but
were either limited by focusing on single types of alterations,
sample size, clinical annotation, or the lack of data
integration.
29,30,40
To overcome these limitations, we integrated significant genetic
alterations composed of recurrent somatic mutations, SCNAs and
SVs, and discovered five genetically distinct DLBCL subsets
29
that
predicted outcome to stateoftheart frontline treatment, suggested
new insights into the lymphomagenesis of DLBCLs and suggested
rational combination treatment (Figure 2). These five subsets
included: (1) a highrisk ABC DLBCLs with nearuniform BCL2 copy
gain, frequent activating MYD88 and CD79B mutations and extra-
nodal tropism (C5 DLBCLs); (2) previously unappreciated favorable
risk ABC DLBCLs with genetic features of an extrafollicular, possibly
marginal zone origin (C1 DLBCLs); (3) poorrisk GCBDLBCLs
with BCL2 SVs, inactivating mutations and/or copy loss of PTEN
and alterations of epigenetic enzymes (C3 DLBCLs); (4) a newly
defined group of goodrisk GCB DLBCLs with distinct alterations in
BCR/phosphoinositide 3kinase (PI3K), JAK/STAT and BRAF
pathway components and multiple histones (C4 DLBCLs); and (5) an
ABC/GCBindependent group of tumors with biallelic inactivation
of TP53,9p21.3/CDKN2A and associated genomic instability
(C2 DLBCLs).
The biology of these five DLBCL subsets was largely confirmed
by an independent nonoverlapping largescale study that followed a
completely different analytical approach and discovered similar
groups with shared pathogenetic mechanisms.
30
In particular, C1
DLBCLs
29
were similar to BN2 (BCL6/NOTCH2),
30
C3 DLBCLs
29
were similar to EZB (EZH2/BCL2)
30
and C5 DLBCLs
29
were similar
to MCD (MYD88/CD79B) (Figure 2).
30
In subsequent work, the
biology of the remaining DLBCL subtypes (C2 and C4 DLBLCs) have
recently been also independently validated by two other groups
(C4 ST2; C2 A53 |C4 SGK1; C2 TP53) (Figure 2)
66,67
.
These studies underscore that DLBCL is a genetically heterogenous
disease with at least 5 different molecular subtypes. At the same
time, it also highlights, that despite having several hundreds of tu-
mors, we still do not capture the full spectrum of DLBCL subsets
and in particular for lowfrequency subtypes larger sample sizes are
needed.
Importantly, the genetic DLBCL subtypes has provided new
insights into the pathogenesis of these tumors with suggestions for
novel rational combination therapies and predict outcome.
29
In a
proofofconcept study, it has been demonstrated that in cell line
models of C3 DLBCLs the combination of BCL2 and PI3Kα/δin-
hibitors cured 75% of the xenotransplanted mice, underscoring
that the new classification is a roadmap to genotypeinformed
targeted combination therapies.
29,68
More work is needed, to link
genetically defined subtypes to actionable vulnerabilities, and
eventually build molecular classifiers that link genetics to de-
pendencies with the hope of using them for precision medical
trials.
WIENAND AND CHAPUY
-
27
6
|
OUTLOOK/PERSPECTIVE
After several decades of empirical clinical trial design, we are reaching
a ceiling for clinical improvements in unselected patients.
69
For that
reason, it is important that technological advances lead to compre-
hensive genomic studies in informative and large LBCL cohorts, that
identify unique genetic signatures for discrete LBCL subtypes. The
derived genetic signatures, their temporal ordering and the associ-
ated mutational signatures and transcriptional profiles have aided in
better understanding of the associated biology and lymphomagenesis,
identified prognostic markers of response to standard induction
treatment and suggested rationally treatment combinations suited for
individual lymphoma subtypes. Importantly, new treatment modal-
ities are quickly moving into the clinics, including molecular subtype
agnostic therapies, such as CART cells, antibody–drug conjugates
and bispecific antibodies, but also other subtypespecific (precision
medicine) approaches. Comprehensive genomic studies suggest that
within one lymphoma entity there are subtypes, which differ geneti-
cally as largely as two unrelated cancer types, while at the same time
there are shared genetic signatures across entities. This information
needs to be appreciated in nextgeneration clinical trial designs (i.e.,
basket vs. umbrella trials). Many driver alterations of LBCL subtypes
encode for proteins that are readily druggable with small molecule
inhibitors or monoclonal antibodies and for LBCL subtypes that are
still poorly understood we need to create model systems, perform
functional screens (both genetic and pharmacological) and integrate
this information with other omics technologies, such as recently
suggested with microenvironmental immune signatures.
70
While the
last years were focused on bulk genomics, recent studies shed light on
the tremendous intratumoral heterogeneity within DLBCL
71
and pave
the way for the coming years. Eventually, molecular classifiers have
the potential to perform patient stratification in realtime and in
clinical settings. We believe that the integration of bulk and singlecell
omics technologies with largescale functional screens will help to
select more specific therapies for patients with highrisk subtypes of
Bcell lymphoma and thereby hopefully, bring precision medicine for
DLBCL to its full potential.
ACKNOWLEDGMENT
KW is supported by the BMBF SeneSys Consortium (031L0189B)
granted towards BC.
CONFLICT OF INTERESTS
The authors declare no potential conflict of interests.
PEER REVIEW
The peer review history for this article is available at https://publons.
com/publon/10.1002/HON.2847.
DATA AVAILABILITY STATEMENT
Not relevant.
REFERENCES
1. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the
World Health Organization classification of lymphoid neoplasms.
Blood. 2016;127(20):23752390.
2. Savage KJ, Monti S, Kutok JL, et al. The molecular signature of
mediastinal large Bcell lymphoma differs from that of other diffuse
large Bcell lymphomas and shares features with classical Hodgkin
lymphoma. Blood. 2003;102(12):38713879.
3. Feuerhake F, Kutok JL, Monti S, et al. NFkappaB activity, function,
and targetgene signatures in primary mediastinal large Bcell lym-
phoma and diffuse large Bcell lymphoma subtypes. Blood.
2005;106:13921399.
4. Rosenwald A, Wright G, Leroy K, et al. Molecular diagnosis of pri-
mary mediastinal B cell lymphoma identifies a clinically favorable
FIGURE 2 Discovery of genetic DLBCL subtypes (A) with associated genetic, transcriptomic and prognostic features (B). DLBCL, diffuse
large Bcell lymphoma
28
-
WIENAND AND CHAPUY
subgroup of diffuse large B cell lymphoma related to Hodgkin lym-
phoma. J Exp Med. 2003;198(6):851862.
5. Green MR, Monti S, Rodig SJ, et al. Integrative analysis reveals se-
lective 9p24.1 amplification, increased PD1 ligand expression, and
further induction via JAK2 in nodular sclerosing Hodgkin lymphoma
and primary mediastinal large Bcell lymphoma. Blood. 2010;
116(17):32683277.
6. Rui L, Emre NC, Kruhlak MJ, et al. Cooperative epigenetic modu-
lation by cancer amplicon genes. Cancer Cell. 2010;18(6):590605.
7. Y Hao, Chapuy B, Monti S, Sun HH, Rodig S, Shipp MA et al., Se-
lective JAK2 inhibition specifically decreases Hodgkin lymphoma
and mediastinal large Bcell lymphoma growth in vitro and in vivo.
Clin Cancer Res. 2014;20(10):2674–83. https://doi.org/10.1158/
10780432.CCR133007
8. Chapuy B, Roemer MG, Stewart C, et al. Targetable genetic features
of primary testicular and primary central nervous system lym-
phomas. Blood. 2016;127(7):869881.
9. Green MR, Rodig S, Juszczynski J, et al. Constitutive AP1 activity
and EBV infection induce PDL1 in Hodgkin lymphomas and post-
transplant lymphoproliferative disorders: implications for targeted
therapy. Clin Cancer Res. 2012;18(6):16111618.
10. Twa DD, Chan FC, BenNeriah S, et al. Genomic rearrangements
involving programmed death ligands are recurrent in primary
mediastinal large Bcell lymphoma. Blood. 2014;123(13):
20622065.
11. Ansell SM, Lesokhin AM, Borrello I, et al. PD1 blockade with
nivolumab in relapsed or refractory Hodgkin's lymphoma. N Engl J
Med. 2014.372(4):311–319. https://doi.org/10.1056/NEJMoa141
1087
12. Younes A, Santoro A, Shipp M, et al. Nivolumab for classical Hodg-
kin's lymphoma after failure of both autologous stemcell trans-
plantation and brentuximab vedotin: a multicentre, multicohort,
singlearm phase 2 trial. Lancet Oncol. 2016;17(9):12831294.
13. Armand P, Shipp MA, Ribrag V, et al. Programmed death1 blockade
with pembrolizumab in patients with classical Hodgkin lymphoma
after brentuximab vedotin failure. J Clin Oncol. 2016;34(31):
37333739.
14. Armand P, Chen YB, Redd RA, et al. PD1 blockade with pem-
brolizumab for classical Hodgkin lymphoma after autologous stem
cell transplantation. Blood. 2019;134(1):2229.
15. Zinzani PL, Ribrag CH, Moskowitz CH, et al. Safety and tolera-
bility of pembrolizumab in patients with relapsed/refractory pri-
mary mediastinal large Bcell lymphoma. Blood. 2017;130(3):
267270.
16. Armand P, Rodig V, Melnichenko C, et al. Pembrolizumab in relapsed
or refractory primary mediastinal large Bcell lymphoma. J Clin
Oncol. 2019;37(34):32913299.
17. Chapuy B SC, Dunford AJ, Kim J, et al. Genomic analyses of PMBL
reveals new drivers and mechanisms of sensitivity to PD1 blockade.
Blood. 2019;134:23692382.
18. Mottok A HSS, Chavez EA, Woolcock B, et al. Integrative genomic
analysis identifies key pathogenic mechanisms in primary medias-
tinal large Bcell lymphoma. Blood. 2019;134:802813.
19. Wienand K, Chapuy B, Stewart C, et al. Genomic analyses of flow
sorted Hodgkin ReedSternberg cells reveal complementary mech-
anisms of immune evasion. Blood Adv. 2019;3(23):40654080.
20. Wang S, Jia M, He Z, Liu XS. APOBEC3B and APOBEC mutational
signature as potential predictive markers for immunotherapy
response in nonsmall cell lung cancer. Oncogene. 2018;37(29):
39243936.
21. Le DT, Uram JN, Wang H, et al. PD1 blockade in tumors with
mismatchrepair deficiency. N Engl J Med. 2015;372(26):25092520.
22. Rubenstein JL, Gupta NK, Mannis GN, Lamarre AK, Treseler P. How
I treat CNS lymphomas. Blood. 2013;122(14):23182330.
23. Cheah CY, Wirth A, Seymour JF. Primary testicular lymphoma. Blood.
2013.123(4):486–493. https://doi.org/10.1182/blood201310
530659
24. Deckert M, MontesinosRongen M, Brunn A, Siebert R. Systems
biology of primary CNS lymphoma: from genetic aberrations to
modeling in mice. Acta Neuropathol. 2014;127(2):175188.
25. Fukumura K, Kawazu M, Kojima T, et al. Genomic characterization of
primary central nervous system lymphoma. Acta Neuropathol
2016;131:865875.
26. Nayak L, Iwamoto FM, LaCasce A, et al. PD1 blockade with nivo-
lumab in relapsed/refractory primary central nervous system and
testicular lymphoma. Blood. 2017;129(23):30713073.
27. Monti S, Chapuy B, Takeyama K, et al. Integrative analysis reveals an
outcomeassociated and targetable pattern of p53 and cell cycle
deregulation in diffuse large B cell lymphoma. Cancer Cell.
2012;22(3):359372.
28. Davis RE, Ngo VN, Lenz G, et al. Chronic active Bcellreceptor
signalling in diffuse large Bcell lymphoma. Nature. 2010;
463(7277):8892.
29. Chapuy B, Stewart C, Dunford AJ, et al. Molecular subtypes of
diffuse large B cell lymphoma are associated with distinct patho-
genic mechanisms and outcomes. Nat Med. 2018;24(5):679690.
30. Schmitz R, Wright GW, Huang DW, et al. Genetics and pathogenesis
of diffuse large Bcell lymphoma. N Engl J Med. 2018;378(15):1396
1407.
31. Phelan JD, Young RM, Webster DE, et al. A multiprotein super-
complex controlling oncogenic signalling in lymphoma. Nature.
2018;560:387391.
32. Thurner L, Preuss KD, Bewarder M, et al. HyperNglycosylated
SAMD14 and neurabinI as driver autoantigens of primary central
nervous system lymphoma. Blood. 2018;132(26):27442753.
33. Chapuy B. Tropism of PCNSL: is it all about autoantigens?. Blood.
2018;132(26):27042706.
34. Lionakis MS, Dunleavy K, Roschewski M, et al. Inhibition of B cell
receptor signaling by ibrutinib in primary CNS lymphoma. Cancer
Cell. 2017;31(6):833843.e835.
35. Grommes C, Pastore A, Palaskas N, et al. Ibrutinib unmasks critical
role of bruton tyrosine kinase in primary CNS lymphoma. Cancer
Discov. 2017;7(9):10181029.
36. Sesques P, Johnson NA. Approach to the diagnosis and treatment of
highgrade Bcell lymphomas with MYC and BCL2 and/or BCL6
rearrangements. Blood. 2017;129(3):280288.
37. Horn H, Ziepert M, Becher C, et al. MYC status in concert with BCL2
and BCL6 expression predicts outcome in diffuse large Bcell lym-
phoma. Blood. 2013;121(12):22532263.
38. Johnson NA, Slack GW, Savage KJ, et al. Concurrent expression of
MYC and BCL2 in diffuse large Bcell lymphoma treated with rit-
uximab plus cyclophosphamide, doxorubicin, vincristine, and pred-
nisone. J Clin Oncol. 2012;30:34523459.
39. Green TM, Young KH, Visco C, et al. Immunohistochemical double
hit score is a strong predictor of outcome in patients with diffuse
large Bcell lymphoma treated with rituximab plus cyclophospha-
mide, doxorubicin, vincristine, and prednisone. J Clin Oncol.
2012;30(28):34603467.
40. Reddy A, Zhang J, Davis NS, et al. Genetic and functional drivers of
diffuse large B cell lymphoma. Cell. 2017;171(2):481494.e15.
41. Rosenwald A, Sehn LH, MaucortBoulch D. Prognostic significance of
<em>MYC</em>single, double, triple hit and <em>MYC</em>
Translocation partner status in diffuse large Bcell lymphoma—a
study by the Lunenburg lymphoma biomarker consortium (LLBC).
Blood. 2018;132(Supp 1):344–344.
42. Shipp MA, Harrington DP, Anderson JR, et al. A predictive model for
aggressive nonHodgkin's lymphoma. N Engl J Med. 1993;329(14):
987994.
WIENAND AND CHAPUY
-
29
43. Sehn LH, Berry B, Chhanabhai M, et al. The Revised International
Prognostic Index (RIPI) is a better predictor of outcome than the
standard IPI for patients with diffuse large Bcell lymphoma treated
with RCHOP. Blood. 2007;109(5):18571861.
44. Hartmann S, Doring C, Vucic E, et al. Array comparative genomic
hybridization reveals similarities between nodular lymphocyte pre-
dominant Hodgkin lymphoma and T cell/histiocyte rich large B cell
lymphoma. Br J Haematol. 2015;169(3):415422.
45. Chen BJ, Chapuy B, Ouyang J, et al. PDL1 expression is charac-
teristic of a subset of aggressive Bcell lymphomas and virusasso-
ciated malignancies. Clin Cancer Res. 2013;19(13):34623473.
46. Griffin GK, Roemer MGM, Lipschitz M, et al. Integrated genetic and
topological analysis reveals a Hodgkinlike mechanism of immune
escape in Tcell/histiocyterich large Bcell lymphoma. Blood.
2018;132(S1):1579–1579.
47. Rosenwald A, Wright G, Chan WC, et al. The use of molecular
profiling to predict survival after chemotherapy for diffuse largeB
cell lymphoma. N Engl J Med. 2002;346(25):19371947.
48. Monti S, Savage KJ, Kutok JL, et al. Molecular profiling of diffuse
large Bcell lymphoma identifies robust subtypes including one
characterized by host inflammatory response. Blood. 2005;105(5):
18511861.
49. Ngo VN, Young RM, Schmitz R, et al. Oncogenically active MYD88
mutations in human lymphoma. Nature. 2011;470(7332):115119.
50. Caro P, Kishan U, Norberg IA, et al. Metabolic signatures uncover
distinct targets in molecular subsets of diffuse large B cell lym-
phoma. Cancer Cell. 2012;22(4):547560.
51. Chen L, Monti S, Juszczynski J, et al. SYK inhibition modulates
distinct PI3K/AKTdependent survival pathways and cholesterol
biosynthesis in diffuse large B cell lymphomas. Cancer Cell.
2013;23(6):826838.
52. Basso K, DallaFavera R. Germinal centres and B cell lymphoma-
genesis. Nat Rev Immunol. 2015;15:172184.
53. Alizadeh AA, Eisen MB, Davis C, et al. Distinct types of diffuse large
Bcell lymphoma identified by gene expression profiling. Nature.
2000;403(6769):503511.
54. Norberg E, Lako A, Chen PH, et al. Differential contribution of the
mitochondrial translation pathway to the survival of diffuse large B
cell lymphoma subsets. Cell Death Differ. 2017;24(2):251262.
55. Hans CP, Weisenburger DD, Greiner TC, et al. Confirmation of
the molecular classification of diffuse large Bcell lymphoma by
immunohistochemistry using a tissue microarray. Blood. 2004;
103(1):275282.
56. Scott DW, Wright GW, Williams PM, et al. Determining celloforigin
subtypes of diffuse large Bcell lymphoma using gene expression in
formalinfixed paraffinembedded tissue. Blood. 2014;123(8):
12141217.
57. Lenz G, Wright G, Dave S, et al. Stromal gene signatures in largeB
cell lymphomas. N Engl J Med. 2008;359(22):23132323.
58. Pfeifer M, Grau M, Lenze D, et al. PTEN loss defines a PI3K/AKT
pathwaydependent germinal center subtype of diffuse large Bcell
lymphoma. Proc Natl Acad Sci U. S. A. 2013;110(30):1242012425.
59. Dubois S, Viailly PJ, Bohers E, et al. Biological and clinical relevance
of associated genomic alterations in MYD88 L265P and nonL265P
mutated diffuse large Bcell lymphoma: analysis of 361 cases. Clin
Cancer Res. 2017;23(9):22322244.
60. Ennishi D, Mottok A, BenNeriah S, et al. Genetic profiling of MYC
and BCL2 in diffuse large Bcell lymphoma determines celloforigin
specific clinical impact. Blood. 2017;129(20):27602770.
61. Lenz G, Staudt LM. Aggressive lymphomas. N Engl J Med.
2010;362:14171429.
62. Lenz G, Wright GW, Emre NC, et al. Molecular subtypes of diffuse
large Bcell lymphoma arise by distinct genetic pathways. Proc Natl
Acad Sci U. S. A. 2008;105(36):1352013525.
63. Pasqualucci L, Trifonov V, Fabbri G, et al. Analysis of the coding
genome of diffuse large Bcell lymphoma. Nat Genet.
2011;43(9):830837.
64. Morin RD, MendezLago M, Mungall AJ, et al. Frequent mutation of
histonemodifying genes in nonHodgkin lymphoma. Nature. 2011;
476(7360):298303.
65. Lohr JG, Stojanov P, Lawrence MS, et al. Discovery and prioritization
of somatic mutations in diffuse large Bcell lymphoma (DLBCL) by
wholeexome sequencing. Proc Natl Acad Sci U. S. A.
2012;109(10):38793884.
66. Wright G HDW, Phelan JD, Coulibaly ZA, et al. A probabilistic
classification tool for genetic subtypes of diffuse large B cell lym-
phoma with therapeutic implications. Cancer Cell. 2020;37:551568.
67. Lacy S SE, Beer BSL, Painter D, et al. Targeted sequencing in DLBCL,
molecular subtypes, and outcomes: a Haematological Malignancy
Research Network Report. Blood. 2020;135(20):17591771.
68. Bojarczuk K, Wienand K, Ryan JA, et al. Targeted inhibition of
PI3Kalpha/delta is synergistic with BCL2 blockade in genetically
defined subtypes of DLBCL. Blood. 2018;133(1):7080.
69. Goy A. Succeeding in breaking the RCHOP ceiling in DLBCL:
Learning from negative trials. J Clin Oncol. 2017;35(31):35193522.
70. Dufva O, Polonen P, Bruck O, et al. Immunogenomic landscape of
hematological malignancies. Cancer Cell. 2020;38(3):380399.e13.
71. Roider T, Seufert J, Uvarovskii A, et al. Dissecting intratumour
heterogeneity of nodal Bcell lymphomas at the transcriptional, ge-
netic and drugresponse levels. Nat Cell Biol. 2020;22(7):896906.
How to cite this article: Wienand K, Chapuy B. Molecular
classification of aggressive lymphomas—past, present, future.
Hematological Oncology. 2021;39(S1):2430. https://doi.org/
10.1002/hon.2847
30
-
WIENAND AND CHAPUY
Received: 3 March 2021
DOI: 10.1002/hon.2860
SUPPLEMENT ARTICLE
Mantle cell lymphoma—Advances in molecular biology,
prognostication and treatment approaches
Elisabeth Silkenstedt |Martin Dreyling
Department of Medicine III, LMU Hospital,
Munich, Germany
Correspondence
Martin Dreyling, Oberarzt der Klinik,
Medizinische Klinik III, LMU Klinikum,
Marchioninistrasse 15, 81377 München,
Germany.
Email: Martin.Dreyling@med.uni-muenchen.de
Funding information
Open Access funding enabled and organized
by Projekt DEAL.
Abstract
Mantle cell lymphoma (MCL) is clinically characterized by its heterogenous behavior
with courses ranging from indolent cases that do not require therapy for years to
highly aggressive MCL with very limited prognosis. A better understanding of the
complex biology of MCL has already led to the approval of several innovative
agents, expanding the landscape of MCL therapies and improving therapeutic op-
tions especially for refractory or relapsed disease. Nevertheless, to further optimize
MCL treatment, early identification of individual risk profile and riskadapted,
patienttailored choice of therapeutic strategy needs to be prospectively incorpo-
rated in clinical patient management. This review highlights recent advances in
deciphering the molecular background of MCL, the definition of prognostically
relevant factors and the identification of potential druggable targets and summa-
rizes current treatment recommendations for primary and relapsed/refractory MCL
including novel targeted therapies.
KEYWORDS
genetics, mantle cell lymphoma, pathogenesis, prognostication, therapy
1
|
INTRODUCTION
Mantle cell lymphoma (MCL) is clinically characterized by its heter-
ogenous behavior with courses ranging from indolent cases that do
not require therapy for years to highly aggressive MCL with very
limited prognosis.
1
Patients typically present with lymphadenopathy
of several sites, most of the patients are diagnosed with advanced
stage disease (Ann Arbor stage III, IV). Extranodal manifestations
occur in 90% of patients, including infiltration of bone marrow (53%–
82%), blood (50%), liver (25%), and the gastrointestinal tract (20%–
60%).
1,2
The spleen is enlarged in 40% of patients.
1
In some cases,
leukemic manifestation in combination with massive splenomegaly is
clinically prominent. These nonnodal, leukemic cases are often
characterized by a more indolent clinical course. Accordingly, in the
WHO 2016 update of lymphoid malignancies, MCL now consists
of two distinct categories.
3
Nodal MCL (80%–90% of cases) is
characterized by unmutated immunoglobulin heavy chain variable
region genes (IGHV), sexdetermining region Ybox 11 (SOX11)
overexpression and a generally more aggressive clinical behavior.
Nonnodal leukemic MCL (10%–20% of cases) typically displays
mutated IGHV, SOX11 negativity and presents with indolent bio-
logical behavior. Histologically, besides “classical” MCL, pleomorphic
and blastoid variants can be distinguished.
3
MCL with blastoid
morphology often features high proliferation rates, displaying a more
aggressive clinical course.
3,4
Traditionally, MCL was associated with a poor prognosis with a
median overall survival (OS) of 3–5 years. However, major advances
in the treatment of MCL patients have been achieved over the last
This is an open access article under the terms of the Creative Commons AttributionNonCommercialNoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is noncommercial and no modifications or adaptations are made.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
Hematological Oncology. 2021;39(S1):3138. wileyonlinelibrary.com/journal/hon
-
31
years, especially by the development of an induction immunoche-
motherapy including cytarabine and antiCD20 antibodies and by
introducing a consolidation highdose therapy with autologous stem
cell transplantation (ASCT).
5
Moreover, introduction of rituximab as
maintenance therapy, especially for those patients not eligible for
highdose therapy, significantly improved survival rates in this group
of patients.
6
Yet, longterm prognosis is still limited and patients with
relapsed/refractory disease usually have a dismal outcome. There-
fore, improved understanding of cellular and molecular biology of
MCL and identification of relevant factors determining prognosis to
optimally use riskadapted treatment approaches will be critical to
further improve outcomes in this disease.
2
|
PATHOGENESIS AND MOLECULAR BIOLOGY
The development of MCL is the result of a complex pathogenetic
interplay between cellular and microenvironmental processes.
Genetic hallmark of MCL and considered the primary oncogenic
event in the pathogenesis is the chromosomal t(11;14)(q13;q32)
translocation, leading to overexpression of cyclin D1 and dysregula-
tion of cell cycle at the G1–S phase transition.
7
Cyclin D1 negative
MCLs usually carry CCND2/CCND3 rearrangements with immuno-
globulin genes instead. A subset of cyclin D1
/D2
/D3
MCL with
aggressive features has cyclin E dysregulation. In these rare cyclin
D1negative cases, immunohistochemistry for SOX11 is especially
helpful to confirm the diagnosis.
The transcription factor SOX11 is overexpressed in more than
90% of MCL cases, whereas a leukemic nonnodal variant, resembling
chronic lymphocytic leukemia (CLL), lacks SOX11 expression and is
associated with a more indolent course.
8
In this subset of patients
with leukemic, nonnodal presentation, SOX11 expression proved to
be prognostically relevant, identifying a favorable outcome in pa-
tients with negative SOX11 with mutated IGHV.
The constitutive activation of the Bcell receptor and its multiple
downstream signaling pathways also plays an important role in the
development of the disease.
7
Furthermore, genomic profiling revealed a high number of sec-
ondary genetic alterations and recurrent mutations affecting for
example regulation of cell cycle, DNA damage response and apoptosis
pathways that contribute to the pathogenesis and aggressiveness of
MCL.
7
In recent years, next generation sequencing approaches to
unravel the genetic background of MCL led to the identification
of numerous recurrent somatic mutations including genes involved
in genotoxic stress pathways (ATM,TP53,CDKN2A), epigenetic
regulators (WHSC1,KMT2D,MEF2B,KMT2C,SMARCA4,SMARCB1)
and genes regulating DNA replication (SAMHD1), RNA processing
(HNRNPH1) as well as cell homeostasis, cell growth and cell
death (CCND1,TP53,CDKN2A,CDKN1B,BIRC3,CARD11,TRAF2,
RB1,POT1,NOTCH1/2).
9
Yet, functional relevance remains un-
clear for most of the mutations and is currently under further
investigation.
3
|
PROGNOSTIC FACTORS
Important clinical and serological factors, associated with a worse
clinical outcome include age, poor general condition, advanced stage
of disease (Ann Arbor stage III or IV), splenomegaly and anemia, the
serum level of β2microglobulin and lactate dehydrogenase (LDH),
blastoid cytology, extranodal presentation, and constitutional
symptoms.
A prognostic score that has been confirmed in numerous series,
the MCL International Prognostic Index (MIPI), was established
implementing four independent prognostic factors: age, performance
status, LDH, and leukocyte count.
10
Yet, the most important prognostic markers independent of
clinical features are the proliferation rate as measured by Ki67
expression and the expression of p53. These two (p53 high and
Ki67 >30%), together with blastoid morphology, were recently
reported to define a highrisk biology with significantly shorter
failurefree and OS. Immunohistochemical determination of Ki67
expression has been prospectively confirmed as a reliable prog-
nostic marker and is, in combination with the MIPI (MIPIc), a
highly recommended tool to estimate individual risk profile and to
identify highrisk patients (Ki67 >30%) who may qualify for more
aggressive therapeutic approaches.
11
Furthermore, a cell prolifer-
ation gene signature (MCL35) that distinguishes patient subsets
that differ by more than 5 years in median survival has been
identified.
Deletions of 17p13 or mutations of TP53 as well as deletions of
CDKN2A were reported to be associated with worse clinical
outcome in the majority of the studies published. Despite treated
with highdose cytarabine and ASCT, younger MCL patients
with deletions of CDKN2A (p16) and TP53 show an unfavorable
prognosis. Furthermore, TP53 mutations were significantly associ-
ated with high Ki67 (>30%), blastoid morphology, MIPI high
risk, and inferior responses to both inductionand highdose
chemotherapy.
12
Other genetic lesions with inferior outcomes include mutations
in the NOTCH genes and in KMT2D as well as MYC alterations and
mutations in WHSC1 and CCND1.
Prognostic markers current and future are summarized in
Table 1.
Concerning the prognostic impact of minimal residual disease
(MRD) status, several studies have been published, providing evi-
dence of the strong prognostic potential of MRD status predicting
improved subsequent progressionfree survival (PFS) for MRD
negative patients at the end of induction and before highdose
consolidation.
5
In addition, lack of molecular remission after end of
currently recommended standard treatment was shown to be
strongly predictive for early clinical relapse within 1–2 years.
5,13
However, so far use of MRD analysis is restricted to clinical trials.
Furthermore, the impact of MRD monitoring in the context of the
newtargeted treatments, such as the Bruton's tyrosine kinase (BTK)
inhibitor ibrutinib, remains unclear.
32
-
SILKENSTEDT AND DREYLING
3.1
|
Therapy in patients 65 years
3.1.1
|
Induction: Doseintensified, cytarabine
containing regimen
The administration of the RCHOP/DHAP regimen compared to
administration of RCHOP alone prior to myeloablative consolida-
tion with ASCT more than doubled time to treatment failure (TTF)
(109 vs. 47 months).
5
We do not recommend an additional routine
central nervous system prophylaxis.
Another commonly used treatment approach, predominantly
applied in the USA, is the intensive chemoimmunotherapy regimen
rituximab in combination with hyperCVAD. This regimen achieved
high complete response (CR) rates and longterm remissions and does
not require consolidation with ASCT. However, this regimen is
hampered by significant therapyassociated toxicity, including sec-
ondary malignancies, and should only be considered in young, fit
patients.
3.1.2
|
Consolidation: ASCT
In several studies, the addition of highdose consolidation followed
by ASCT resulted in impressive survival rates.
5,14
A large randomized
trial proved that consolidation by myeloablative radiochemotherapy
followed by ASCT in first remission significantly prolonged PFS (3.3
vs. 1.5 years) and OS
15
independently of the addition of rituximab.
Unfortunately, even after such intensive consolidation regimen, a
majority of patient's relapse.
3.1.3
|
Maintenance
Rituximab maintenance after ASCT is currently considered standard
of care for younger patients with MCL based on the results of a large
phase III trial showing a significant optimization of PFS (83% vs. 64%
after 4 years) and OS (89% vs. 80% after 4 years) after 3 years of
rituximab maintenance compared to observation only.
14
Recently, another phase III trial revealed a benefit from a
lenalidomide maintenance after autologous transplantation with
improved PFS (80% vs. 64% after 3 years) compared to observa-
tion.
16
However, due to the elevated toxicity profile (especially
hematotoxicity), lenalidomide maintenance should be only applied to
patients not suitable to receive rituximab.
Figure 1suggests a riskadapted treatment strategy for patients
65 years.
3.2
|
Therapy in patients more than 65 years
3.2.1
|
Induction
The group of the over 65 year olds ineligible for transplantation
presents very heterogenous regarding physical and cognitive per-
formance. A suggested therapeutic algorithm is depicted in Figure 2.
Fit patients greater than 65 years should receive conventional
immunochemotherapy followed by rituximab maintenance.
6
The VR
CAP regimen recently proved to be superior over RCHOP in a large
International phase III trial with a doubled OS after 82 months (90.7
vs. 45.7 months). However, hematologic toxicity (especially >Grade 3
thrombopenia) was significantly increased in the experimental arm
(57% vs. 6%).
17
The RBAC scheme offers another useful option. Yet,
this regimen was accompanied by severe hepatotoxicities and should
therefore only be administered to very fit older patients with high
risk features (e.g., blastoid variant, high LDH count).
18
Alterna-
tively, for patients not qualifying for such intensive therapy regi-
mens, RBendamustine offers an appropriate alternative. This
combination resulted in similar response rates (93% vs. 91%)
compared to RCHOP and was even superior in PFS (35 vs. 21
months) with a more favorable toxicity profile observed.
19
Taken
TABLE 1Prognostic markers—
current and future In clinical routine Potential for future use
Age “MCL35” RNA expression analysis
Performance status SOX11 expression
Central nervous system involvement at diagnosis TP53 mutations/deletions by sequencing
analysis
Stage of disease (I and II vs. III and IV) MRD testing
Serum level of β2microglobulin and LDH
Morphology (classic vs. blastoid)
MIPI
Ki67 (<30% vs. >30%)
TP53 expression by immunohistochemistry
Abbreviations: LDH, lactate dehydrogenase; MIPI, MCL International Prognostic Index; MRD,
minimal residual disease.
SILKENSTEDT AND DREYLING
-
33
FIGURE 1 Suggested therapeutic algorithm
34
-
SILKENSTEDT AND DREYLING
together, VRCAP and bendamustine–rituximab (BR) represent the
current standard approaches in older patients not eligible for high
dose therapy, who represent the majority of MCL patients. VR
CAP should be preferably considered for patients with a higher
riskprofile such as high Ki67 expression or blastoid morphology. BR
may be preferable especially in patients with a more indolent CLL
like presentation.
3.2.2
|
Maintenance
A large, randomized, European phase III trial compared rituximab
maintenance to interferon (IFN) maintenance, confirming superiority
of rituximab as maintenance therapy. In this study, after four years,
58% of the patients receiving rituximab after induction therapy with
RCHOP were in remission, compared to 29% in the IFN arm
(p=0.01). PFS and OS were also significantly improved in the Rit-
uximab arm (5years PFS R vs. IFN 51% vs. 22%, 5 years OS R vs. IFN
79% vs. 59%).
20
Based on these results, rituximab maintenance after
RCHOP is now generally recommended. Although investigation of
the additional benefits of rituximab maintenance therapy after BR
chemotherapy is pending, we recommend a similar approach.
4
|
RECURRENT AND REFRACTORY DISEASE
4.1
|
Moleculartargeted therapies
Several targeted therapy approaches have been investigated in
different studies as single agents or in combination with immu-
nochemotherapies or other targeted therapies (Table 2).
Targeting the Bcell receptor pathway with the BTK inhibitor
ibrutinib resulted in the highest response rates of all targeted ap-
proaches so far leading to its approval in relapsed MCL. In a large
International phase II study, response rates of 68% were achieved
with ibrutinib in patients with relapsed disease.
21
The combination
FIGURE 2 Suggested therapeutic algorithm for patients greater than 65 years
SILKENSTEDT AND DREYLING
-
35
with rituximab was effective in all cases with low Ki67, whereas in
highly proliferating disease, only half of the patients responded to
this approach.
22
A pooled analysis of the results of three different
trials testing ibrutinib as monotherapy revealed overall response
rates of 66% with median PFS and OS of 12.8, respectively 25
months.
23
Ibrutinib given in combination with bendamustine and
rituximab in patients 65 years of age or older with newly diagnosed
MCL is currently being evaluated in the phase 3 SHINE trial
(NCT01776840). However, interindividual responsiveness is heter-
ogenous and primary and secondary resistance has been reported
with poor clinical outcome.
24
In patients with mutations in the P53
gene, median PFS was shown to be significantly worse. Patients
suffering early relapses after ibrutinib therapy demonstrated very
aggressive clinical courses.
24
Second generation BTK inhibitor acalabrutinib was approved in
October 2017 by the Food and Drug Administration (FDA) for pa-
tients with relapsed/refractory MCL who had received at least one
prior therapy as promising results, especially regarding tolerability,
were observed in an openlabel phase 2 study.
25
Acalabrutinib in
combination with BR compared to BR alone in previously untreated
MCL patients more than 65 years of age is currently evaluated in an
ongoing phase3 study (NCT02972840).
Nextgeneration BTKinhibitor zanubrutinib is a highly potent, se-
lective, bioavailable, and irreversible BTK inhibitor with maximized
BTK occupancy. It was approved in 2019 in the United States and
China for the treatment of patients with R/R MCL based on results
from a phase II study in Chinese patients with R/R MCL reporting
high overall response rates with durable CRs and improved safety
and tolerability over existing treatments. The potential for use of
zanubrutinib in the firstline setting is currently under evaluation in
the randomized phase III MANGROVE study (NCT04002297) in
which patients with treatmentnaive MCL will receive zanu-
brutinib +rituximab or bendamustine +rituximab.
For patients suffering early relapses after ibrutinib therapy, a
monotherapy with the Bcelllymphoma 2inhibitor Abt199 (ven-
etoclax) might be a promising alternative, as a phase I trial showed
response rates of 75% in patients with relapsed MCL and 60% in
patients having received prior ibrutinib therapy. Recently, the com-
bination of ibrutinib and venetoclax proved to be highly effective in a
small study cohort.
26
The potential advantage of ibrutinib combined
with venetoclax over ibrutinib alone is currently being examined in
an ongoing phase3 study (SYMPATICO) (NCT03112174).
Various studies confirmed a benefit of the orally available
immunomodulatory drug lenalidomide in relapsed MCL, with response
rates of 35%–50%. In a randomized phase II trial, this approach was
superior to monochemotherapy (response rate 46% vs. 23%). Based
on an in vitro synergism, lenalidomide in combination with rituximab
resulted in long lasting remissions in firstline therapy of a rather low
risk patient cohort.
27
This approach may be applied especially in
patients with contraindications against BTK inhibitors.
4.2
|
Allogeneic transplantation versus CAR Tcells
For younger highrisk patients with TP53—mutated and relapsed MCL,
who are transplant—eligible, the option of allogeneic transplantation
TABLE 2Molecular targeted therapies in MCL
Regimen Phase Number of patients ORR (CR) % Median PFS (months)
Lenalidomide Phase II 134 28 (7.5 ) 4
Lenalidomide Phase II 57 35
12
8.8
Lenalidomide versus. Phase II 170 46
11
8.7
Monochemotherapy 84 23
8
5.2
Lenalidomid +rituximab Phase II 44 57 (36) 11.1
Lenalidomid +rituximab Phase II 38 64% (After 5 years)
Ibrutinib Phase II 111 68
21
13.9
Ibrutinib vs. temsirolimus Phase III 280 72
19
14.6
40
1
6.2
Ibrutinib +rituximab Phase II 50 88 (44)
Ibrutinib +lenalidomid +rituximab Phase II 50 76 (56)
Idelalisib Phase I 40 40
5
3.7
Abt199 (venetoclax) Phase I 28 75
21
14
Abt199 (venetoclax) +ibrutinib Phase II 24 71
Acalabrutinib Phase II 124 81 (40)
Zanubrutinib Phase II 86 84 (68.6) 22.1
Abbreviations: CR, complete response; MCL, mantle cell lymphoma; ORR, overall response rate; OS, overall survival; PFS, progressionfree survival.
36
-
SILKENSTEDT AND DREYLING
could be considered. Reducedintensity allogeneic stem cell trans-
plantation resulted in longterm diseasefree survival in about 30% of
the patients and may be applicable also in patients older than age 60
years. Yet, transplantationassociated severe acute and delayed tox-
icities, including chronic graft versus host disease and 20%–25%
treatmentrelated mortality are frequent. Therefore, allogeneic
transplantation is not recommended in the firstline setting and should
only be discussed in relapsed disease.
1
Recently, the FDA has approved the autologous CD19 CAR Tcell
construct brexucabtagene autoleucel (formerly KTEX19; Tecartus),
based on results of the ZUMA2 trial, evaluating safety and efficacy of
this CD19 CAR Tcell construct in patients with R/R MCL. Treatment
with brexucabtagene autoleucel resulted in the induction of durable
remissions, although serious and lifethreatening toxic events were
reported.
28
Another CD19directed CAR Tcell product (Lisocabta-
gene Maraleucel) for relapsed/refractory MCL is currently being
evaluated in the ongoing Phase 1 study TRANSCEND NHL 001
(NCT02631044). Overall, results are promising and CAR Tcell con-
structs may have also the potential to cure MCL patients. Yet, much
longer followup is needed.
ACKNOWLEDGMENTS
Open Access funding enabled and organized by Projekt DEAL.
CONFLICT OF INTEREST
MD received (institutional) research support from Abbvie, Bayer,
Celgene, Janssen, Roche and honoraria for scientific advisory
boards and speaker's honoraria from Amgen, Astra Zeneca,
Bayer, Beigene, Celgene, Genmab, Gilead, Incyte, Janssen, Novartis,
Roche.
REFERENCES
1. Dreyling M, Campo E, Hermine O, et al. Newly diagnosed and
relapsed mantle cell lymphoma: ESMO clinical practice guidelines for
diagnosis, treatment and followup. Ann Oncol. 2017;28(suppl_4):
iv62iv71.
2. Romaguera JE, Medeiros LJ, Hagemeister FB, et al. Frequency of
gastrointestinal involvement and its clinical significance in mantle
cell lymphoma. Cancer. 2003;97(3):586591.
3. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of
Tumours of Haematopoietic and Lymphoid Tissues. Vol. 2, 4th ed. Lyon,
France: IARC Publications; 2017.
4. Hoster E, Rosenwald A, Berger F, et al. Prognostic value of Ki67 index,
cytology, and growth pattern in mantlecell lymphoma: results from
randomized trials of the European Mantle Cell Lymphoma Network. J
Clin Oncol. 2016;34(12):13861394.
5. Hermine O, Hoster E, Walewski J, et al. Addition of highdose
cytarabine to immunochemotherapy before autologous stemcell
transplantation in patients aged 65 years or younger with mantle
cell lymphoma (MCL Younger): a randomised, openlabel, phase 3
trial of the European Mantle Cell Lymphoma Network. Lancet.
2016;388(10044):565575.
6. KluinNelemans HC, Hoster E, Hermine O, et al. Treatment of older
patients with mantlecell lymphoma. N Engl J Med. 2012;367(6):
520531.
7. Jares P, Colomer D, Campo E. Molecular pathogenesis of mantle cell
lymphoma. J Clin Invest. 2012;122(10):34163423.
8. Fernandez V, Salamero O, Espinet B, et al. Genomic and gene
expression profiling defines indolent forms of mantle cell lymphoma.
Cancer Res. 2010;70(4):14081418.
9. Hill HA, Qi X, Jain P, et al. Genetic mutations and features of mantle
cell lymphoma: a systematic review and metaanalysis. Blood Adv.
2020;4(13):29272938.
10. Hoster E, Dreyling M, Klapper W, et al. A new prognostic index
(MIPI) for patients with advancedstage mantle cell lymphoma.
Blood. 2008;111(2):558565.
11. Aukema SM, Hoster E, Rosenwald A, et al. Expression of TP53 is
associated with the outcome of MCL independent of MIPI and Ki67
in trials of the European MCL Network. Blood. 2018;131(4):
417420.
12. Eskelund CW, Dahl C, Hansen JW, et al. TP53 mutations identify
younger mantle cell lymphoma patients who do not benefit
from intensive chemoimmunotherapy. Blood. 2017;130(17):1903
1910.
13. Pott C, Hoster E, DelfauLarue MH, et al. Molecular remission is an
independent predictor of clinical outcome in patients with mantle
cell lymphoma after combined immunochemotherapy: a European
MCL intergroup study. Blood. 2010;115(16):32153223.
14. Le Gouill S, Thieblemont C, Oberic L, et al. Rituximab after autolo-
gous stemcell transplantation in mantlecell lymphoma. N Engl J
Med. 2017;377(13):12501260.
15. Zoellner A, Unterhalt M, Stilgenbauer S, et al. Autologous stem cell
transplantation in first remission significantly prolongs progres
sionfree and overall survival in mantle cell lymphoma. Interna-
tional Conference on Malignant Lymphoma; 2019; Lugano,
Switzerland.
16. Ladetto M, Ferrero S, Evangelista A, et al. Lenalidomide main-
tenance after autologous transplantation prolongs PFS in young
MCL patients: results of the randomized phase III MCL 0208
trial from Fondazione Italiana Linfomi (FIL). Blood.
2018;132:401.
17. Robak T, Jin J, Pylypenko H, et al. Frontline bortezomib, rituximab,
cyclophosphamide, doxorubicin, and prednisone (VRCAP) versus
rituximab, cyclophosphamide, doxorubicin, vincristine, and predni-
sone (RCHOP) in transplantationineligible patients with newly
diagnosed mantle cell lymphoma: final overall survival results of a
randomised, openlabel, phase 3 study. Lancet Oncol. 2018;19(11):
14491458.
18. Visco C, Chiappella A, Nassi L, et al. Rituximab, bendamustine, and
lowdose cytarabine as induction therapy in elderly patients with
mantle cell lymphoma: a multicentre, phase 2 trial from Fondazione
Italiana Linfomi. Lancet Haematol. 2017;4(1):e15e23.
19. Rummel MJ, Niederle N, Maschmeyer G, et al. Bendamustine plus
rituximab versus CHOP plus rituximab as firstline treatment for
patients with indolent and mantlecell lymphomas: an openlabel,
multicentre, randomised, phase 3 noninferiority trial. Lancet.
2013;381(9873):12031210.
20. KluinNelemans HC, Hoster E, Hermine O, et al. Treatment of older
patients with mantle cell lymphoma: longterm followup of the
randomized European MCL Elderly trial. J Clin Oncol. 2020;38(3):
248–256.
21. Wang ML, Rule S, Martin P, et al. Targeting BTK with ibrutinib in
relapsed or refractory mantlecell lymphoma. N Engl J Med. 2013;
369(6):507516.
22. Wang ML, Lee H, Chuang H, et al. Ibrutinib in combination with
rituximab in relapsed or refractory mantle cell lymphoma: a
singlecentre, openlabel, phase 2 trial. Lancet Oncol. 2016;17(1):
4856.
23. Rule S, Dreyling M, Goy A, et al. Ibrutinib for the treatment of
relapsed/refractory mantle cell lymphoma: extended 3.5year
follow up from a pooled analysis. Haematologica. 2019;104(5):
e211e214.
SILKENSTEDT AND DREYLING
-
37
24. Martin P, Maddocks K, Leonard JP, et al. Postibrutinib outcomes
in patients with mantle cell lymphoma. Blood. 2016;127(12):
15591563.
25. Wang M, Rule S, Zinzani PL, et al. Acalabrutinib in relapsed or re-
fractory mantle cell lymphoma (ACELY004): a singlearm, multi-
centre, phase 2 trial. Lancet. 2018;391(10121):659667.
26. Tam CS, Anderson MA, Pott C, et al. Ibrutinib plus Venetoclax for
the treatment of mantlecell lymphoma. N Engl J Med. 2018;378(13):
12111223.
27. Ruan J, Martin P, Christos P, et al. Fiveyear followup of lenalido-
mide plus rituximab as initial treatment of mantle cell lymphoma.
Blood. 2018;132(19):20162025.
28. Wang M, Munoz J, Goy A, et al. KTEX19 CAR Tcell therapy in
relapsed or refractory mantlecell lymphoma. N Engl J Med. 2020;
382(14):13311342.
How to cite this article: Silkenstedt E, Dreyling M. Mantle cell
lymphoma—Advances in molecular biology, prognostication
and treatment approaches. Hematological Oncology.
2021;39(S1):3138. https://doi.org/10.1002/hon.2860
38
-
SILKENSTEDT AND DREYLING
DOI: 10.1002/hon.2856
SUPPLEMENT ARTICLE
Personalized medicine for Hodgkin lymphoma: Mitigating
toxicity while preserving cure
Jemma Longley |Peter W. M. Johnson
Centre for Cancer Immunology, CRUK
Research Centre, University of Southampton,
Southampton, UK
Correspondence
Peter W. M. Johnson, Cancer Research UK
Centre, Southampton General Hospital,
Southampton SO16 6YD, UK.
Email: johnsonp@soton.ac.uk
Funding information
Cancer Research UK
Abstract
The treatment of classical Hodgkin lymphoma in young patients is one of the success
stories of modern medicine. The use of riskand responseadapted approaches to
guide treatment decisions has led to impressive cure rates while reducing the long
term toxicity associated with more intensive therapies. Tissue biomarkers have not
yet proven more effective than clinical characteristics for risk stratification of patients
at presentation, but functional imaging features such as metabolic tumor volume may
be used to predict response, if early observations can be validated. The success of
treatment in younger patients has unfortunately not been mirrored in those over 60,
where complex decisionmaking is often required, with a paucity of data from clinical
trials. The use of PD1 blocking antibodies and brentuximab vedotin in this cohort,
either alone or in combination with chemotherapy, may provide attractive options.
The incorporation of frailty assessment, qualityoflife outcomes, and specialist
geriatric input is also important to ensure the best outcomes for this diverse group.
KEYWORDS
antibody–drug conjugate, checkpoint blocking antibody, FDGPET, Hodgkin lymphoma
1
|
INTRODUCTION
Classical Hodgkin lymphoma (cHL) is a malignancy of germinal center
B cells, characterized by the presence of the Hodgkin Reed–
Sternberg (HRS) cell. It is the most common lymphoid malignancy
diagnosed in children and young adults in developed countries, with a
bimodal distribution peaking in the 2nd and 7th decades. Treatment
with radiotherapy, multimodality chemotherapy regimens, newer
antibody–drug conjugates, and immunotherapy checkpoint inhibitors
(ICI), using 2(
18
F)fluoro2deoxyDglucose (FDG) positron emis-
sion tomography (FDGPET) to direct therapy has translated into 10
year overall survival (OS) rates above 80%.
1
However, the ongoing
challenge remains of identifying patients with highrisk disease who
will benefit the most from intensified therapy, while deescalating
treatment in those likely to be cured by less toxic regimens, to
minimize the longterm morbidity and mortality seen in a minority of
survivors, without comprising outcomes. This article will outline the
emergence of new biomarkers to aid risk stratification and guide
treatment decisions at diagnosis, the use of different response
adapted approaches, and the incorporation of new targeted agents
in the treatment of both younger and older patients, in a more
personalized approach to therapy.
2
|
APPROACHES IN YOUNGER PATIENTS WITH
HODGKIN LYMPHOMA
The initial treatment of advancedstage HL with either bleomycin,
etoposide, doxorubicin, cyclophosphamide, vincristine, procarbazine
and prednisolone (escalated BEACOPP) or doxorubicin, vinblastine,
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, pro-
vided the original work is properly cited.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
Hematological Oncology. 2021;39(S1):3945. wileyonlinelibrary.com/journal/hon
-
39
bleomycin, and dacarbazine (ABVD) is guided by risk stratification at
diagnosis, comorbidity, and patient preference. Excellent disease
control is achieved with more intensive BEACOPP regimens; how-
ever, this is at the price of increased acute toxicity and longterm
morbidity, including second malignancies, infertility and cardiovas-
cular disease among survivors, when compared to less intensive
regimens. The use of six cycles of ABVD compared to four cycles of
escBEACOPP plus two cycles of standard BEACOPP in the Italian
HD2000 study showed no difference in 10year OS, despite a sig-
nificant difference in progressionfree survival (PFS) in favor of the
BEACOPP group at 5 years.
1
This is perhaps explained by the
significantly lower rates of second malignancy in patients treated
with ABVD had when compared to escBEACOPP (0.7% vs. 6.6%) and
the success of autologous stem cell transplant (ASCT) in the salvage
of relapsed disease. In the modern era, morbidity is predicted to be
lower as the number of BEACOPP cycles has been reduced using a
PETdirected approach
2
; however, identification of patients with
higher risk disease at diagnosis seems important for choosing the
correct intensity of therapy, to optimize the chance of cure.
Risk stratification of patients' disease by the International Prog-
nostic Score (IPS) has previously been used to guide clinicians with
initial treatment decisions; however, compared to the dynamic
assessment of response by PET, it is less able to identify those pa-
tients with highrisk disease that have a poorer outlook.
3
The incor-
poration of biologic features such as gene expression profiles in
addition to IPS has so far not yielded any prospectively validated
biomarkers, but measurement of metabolic tumor volume (MTV) and
total lesional glycolysis (TLG) at the baseline PET may provide a more
quantifiable assessment of tumor burden, a known predictor of poor
outcome.
4
The European collaborative group retrospectively
analyzed baseline total MTV (TMTV) in 258 patients with earlystage
HL in the standard combined modality arm of the H10 trial and
showed that both TMVT and interim PET (iPET) following two cycles
of ABVD were independently prognostic of response to treatment,
and when combined, allowed identification of a highrisk patient
group with a 5year PFS of only 25% (TMVT >148 cm
3
and iPET
positive—Deauville Score [DS] 4–5).
5
In this study, the TMTV was
calculated by summing all the extranodal and nodal lesions using the
41% maximum standardized uptake value threshold (SUVmax)
method. In advancedstage disease (stage IIB–IV), 848 patients
enrolled in the RATHL trial had baseline total/bulk MTV and TLG
measured using SUV 2.5 when compared to the liver (the 41%
SUVmax method was found not to be associated with PFS or 3year
HL events in this patient cohort).
6
Patients with a positive iPET
following two cycles of ABVD had a significantly higher total/bulk
MTV and TLG when compared to iPETnegative patients (p=0.0002);
however, in a multivariate analysis, only total TLG, B symptoms, and
age were significantly associated with PFS. Patients with a negative
iPET and highvolume TLG at baseline (defined as >3318 g) had a 5
year treatment failure rate of 31%, compared with 13.1% in low
volume TLG.
6
A study which retrospectively analyzed a total of 392
patients enrolled in both arms of the AHL 2011 LYSA trial identified a
small number of patients with a highbaseline PET TMVT (set at a
threshold of 350 ml using the 41% SUVmax method) who had a
positive iPET (DS 4–5) following two cycles of escBEACOPP, with a 2
year PFS of 61% compared to 88% and 96% in patients with a low
TMTV/positive iPET and a low TMTV/negative iPET, respectively.
7
The rate of progression among patients with stage IV disease and a
negative iPET in the RATHL trial was 20% compared with less than
10% of patients enrolled in the GHSG H18 trial and LYSA study,
suggesting a more reliable negative predictive value of iPET after
more intensive regimens such as escalated BEACOPP in patients with
highrisk disease.
3
Thus, baseline total MTV and TLG may prove
useful in the context of guiding initial intensity of treatment, by
identifying those at risk of treatment failure despite a negative iPET.
Measurement of total MTV/TLG will require standardization; how-
ever, similar to the Deauville scoring system that was developed for
iPET assessment to allow reproducibility and consistency when
stratifying patients into different risk groups and setting consistent
threshold values.
4
Prospective validation of this potential biomarker
in a large clinical trial is needed to ascertain its true prognostic value.
Patients with advancedstage HL and a positive iPET after two
cycles of ABVD in the RATHL trial went on to receive escalated
treatment with more intensive BEACOPP regimes (four cycles of
escBEACOPP or six cycles of BEACOPP14), with a 5year PFS of
65.7% and OS of 85.1%. This compares favorably to continuation of
ABVD following iPET in previous studies, where the PFS was consis-
tently less than 40%.
3
The South West Oncology Group (SWOG) 0816
trial showed at 5year followup, 59 patients with advancedstage HL
(here defined as stage III–IV) and a positive iPET (DS 4–5) escalated to
escBEACOPP after two cycles of ABVD had a similar PFS of 66%, but
the rate of second malignancy was 14% with a short median onset of
4.2 years. In this study, six cycles of escBEACOPP were given
compared to four cycles in RATHL, which may partly explain the high
rate of secondary malignancy.
8
The GHSG H18 trial showed that pa-
tients with iPETpositive disease following two cycles of escBEACOPP
who were treated with a total of six cycles of escBEACOPP had a
secondary malignancy rate of 9% at 5.5 years of followup.
2
In the RATHL study, the treatment failed despite escalation to
BEACOPP regimens in 20 out of 37 patients with a DS of 5 on iPET,
and this group almost certainly requires a different approach to
improve their survival. The use of salvage therapy with highdose
chemotherapy (HDT) followed by ASCT is an option for patients
with initial chemorefractory disease, and was investigated by the
Italian HD0801 trial.
9
Here a positive iPET was defined as a DS of 3–
5, and therefore included a more favorable patient group when
compared to outcomes from RATHL and LYSA trials. Following two
cycles of ABVD, 81 (19%) patients remained iPET positive and
received HDT ASCT, with a 2year PFS of 75% suggesting that early
intensification might improve outcomes for this group.
9
The use of
newer agents such as brentuximab vedotin (BV) and antiPD1 anti-
bodies in the frontline treatment of patients with highrisk iPET
positive disease may provide an alternative to ASCT, given their
activity in the relapsed/refractory disease; however, there is as yet
little data to support their use in a PETdriven approach for this
selected group of patients. The Phase III ECHELON1 trial
40
-
LONGLEY AND JOHNSON
incorporated six cycles of brentuximab with AVD chemotherapy
(A +AVD) and showed a 3year modified PFS (including a DS 3–5 at
the end of treatment as an event) of 83.1% compared with 76.2% in
patients' receiving six cycles of ABVD (7.1% difference p=0.005),
with a beneficial trend observed in iPETpositive patients <60 years
receiving A+AVD (3year PFS 69.2% vs. 54.7%, respectively).
10
Therefore, A+AVD may be an attractive option for those patients
with highrisk disease who wish to reduce the risk of longterm
toxicity associated with BEACOPP regimes or who are unable to
tolerate escalation of therapy following a positive iPET.
In early stage unfavorable disease, the addition of BV to four
cycles of AVD within in a phase II PETdirected pilot study in the
United States allowed the reduction of dose and intensity of radio-
therapy without apparently compromising treatment efficacy, with a
2year PFS of 97% among 29 patients who did not receive any
consolidation radiotherapy.
11
The phase III GHSG H17 trial in a
similar patient cohort also showed that the omission of radiotherapy
in those with a negative PET following two cycles of ABVD plus two
cycles of escBEACOPP was noninferior in terms of 5year PFS (2.2%
difference in favor of the radiotherapy group).
12
An initial high
intensity approach in the earlystage disease thus appears to maxi-
mize cure rates without the need for consolidation radiotherapy in
those patients with a negative PET at the end of the treatment,
showing an improvement in the negative predictive value of iPET
when compared to the use of less intensive regimens.
The RATHL trial showed that the omission of bleomycin in patients
with a complete metabolic response at iPET did not compromise sur-
vival outcomes, and resulted in a lower incidence of pulmonary toxicity
(5year PFS and OS 84% and 98% vs. 86% and 97%, respectively).
3
Similarly in the AHL 2011 LYSA trial, 5year PFS was not significantly
different between patients treated with continued escBEACOPP or
deescalated to ABVD (86.2% standard arm vs. 85.7% PETdriven arm)
leading to the conclusion that therapy can be reduced in those patients
whose disease responds to initial therapy without compromising sur-
vival outcomes.
13
The optimal number of escBEACOPP cycles was
investigated by the GHSG H18 trial in this context, and showed that in
patients with a negative iPET (DS 1–2) following two cycles of esc-
BEACOPP, the duration could be safely reduced to two further cycles,
with a small but statistically significant improvement in 5year survival
outcomes when compared to four cycles (PFS 92.2% vs. 90.8% OS 97.7
vs. 95.4%, respectively).
2
For patients with an IPS score of 1–2 and
favorable baseline characteristics, an initial two cycles of ABVD with
deescalation to AVD if iPET negative and escalation to four cycles of
escBEACOPP if iPET positive has a high probability of cure while
minimizing the number of patients exposed to the acute and longterm
toxicity of BEACOPP regimes. The omission of radiotherapy in those
patients with a complete metabolic response did not affect survival
outcomes in the GHSG H15 study
14
and only 6.5% of patients received
consolidation radiotherapy without loss of disease control in the
RATHL trial.
3
There may be a role for radiotherapy in singlesite iPET
positive disease to reduce the number of patients escalated to more
intensive chemotherapy regimens; however, there is currently a lack of
prospective data supporting this approach.
3
|
IMMUNE CHECKPOINT INHIBITORS AND
EMERGING BIOMARKERS
The use of immune checkpoint inhibitors (ICI) in relapsed Hodgkin
lymphoma is well established, and the use of antiPD1 antibodies
combined with multiagent chemotherapy is being explored in the
firstline setting. A study of affected nodes in those treated with anti
PD1 antibodies showed modification of the HL microenvironment in
response to antiPD1 therapy, with rapid depletion of HRS cells and a
reduction in PDL1expressing tumorassociated macrophages and
regulatory T cells.
15
There was no clonal expansion and activation of
cytotoxic T cells as is seen in solid tumors, suggesting a mechanism of
action that is particular to HL, involving interruption of T cell–B cell
signaling pathways. Combination of nivolumab with AVD chemo-
therapy (N +AVD) for advancedstage HL (stage IIB–IV) was inves-
tigated by Ramchandren et al. who first gave nivolumab monotherapy
for 4 doses, followed by combination therapy (N +AVD) for 12 doses
every 2 weeks, with response assessment at the end of monotherapy,
after two combination cycles and at the end of the therapy.
16
Interestingly, at the end of monotherapy, the complete response rate
was 21%, with all patients in the highest quartile for expression of
PDL1 on HRS cells achieving a CR after combination therapy, main-
tained at 32 weeks of followup. Discontinuation rates were low
(10%) with a febrile neutropenia rate of 10%. The most common
endocrine immunemediated adverse event (IMAE) was hypothy-
roidism, and the main nonendocrine IMAE was rash (grade 1–2).
Generally, the regimen was well tolerated, but there was one
treatmentrelated death in an older patient, in CR after two cycles of
combination therapy who experienced four grade 3–4 adverse
events.
16
The use of pembrolizumab monotherapy prior to 4–6 cycles
of AVD in 30 patients with early unfavorable and advanced disease
showed an impressive 100% CR rate by the end of two cycles of AVD.
Responses were durable, with no progression or death at 22 months
of followup, with no consolidation radiotherapy given at the end of
the treatment.
17
Phase III trial data comparing N +AVD versus
BV +AVD in the firstline setting are awaited.
The pattern of disease response in the context of antiPD1
therapy has prompted revision of the Lugano Classification lym-
phoma response criteria, to include immunomodulatory therapy
(LYRIC), due to early imaging suggestive of progressive disease (PD)
in patients who later gained clinical benefit.
18
The phenomenon of
tumor flare or pseudoprogression is well documented in patients
with solid tumors treated with ICI, as a result of immune cell infil-
tration or the delayed effect of these drugs allowing early tumor
growth. The use of early iPET to guide responseadapted treatment
in HL may be particularly difficult to interpret in this context, with
the risk of tumor flare interpreted as a positive iPET and patients
subsequently escalated to more intensive regimes and exposed to
unnecessary toxicity whose disease may have responded at a later
time point. This resulted in the addition of indeterminate response
(IR) to CR PR and PD and allows the flexibility for patients to
continue treatment with further imaging at 12 weeks to confirm
either PD or response (Table 1). There may be a role for antiPD1
LONGLEY AND JOHNSON
-
41
therapy in those patients with a DS of 5 on iPET as an alternative to
escalation of therapy to more intensive regimes, whose disease is
refractory to traditional chemotherapy.
4
|
APPROACHES FOR OLDER PATIENTS WITH
HODGKIN LYMPHOMA
The use of BV and ICI in the elderly may be an attractive option
as monotherapy, or in combination with less toxic chemotherapy
regimens, to improve the poorer survival outcomes when
compared to the younger population. The problems of comorbidity,
poor performance status (PS), increased adverse events, and low
tolerance of chemotherapy regimens at full dose in this hetero-
geneous population have resulted in the reported 3year PFS and
OS rates of 55% and 78%, respectively.
19
Evens et al. investigated
BV in sequential combination with AVD in 48 patients over 60
with untreated HL (stage II–IV) in a Phase II trial, which showed
encouraging 2year PFS and OS of 84% and 93%, respectively.
20
Patients were given two cycles of leadin and consolidation BV,
based on previous studies which have shown poor durability of
responses in older patients treated with BV monotherapy or BV
plus dacarbazine.
21
Only 52% of patients completed the full course
TABLE 1Lymphoma response to immunomodulatory therapy
criteria
IR criteria
IR1 Increase in overall tumor burden 50% of up to six measurable
lesions within the first 12 weeks of treatment without clinical
deterioration
IR2 Appearance of new lesions or increase in one or more existing
lesions 50% at any time during treatment in the context of
<50% increase in overall tumor burden
IR3 Increase in FDG update of 1 lesion(s) without an increase in size
or number of lesions
Abbreviations: FDG, fluoro2deoxyDglucose; IR, indeterminate
response.
TABLE 2List of instrumental activities of daily living
1 Planning, preparing and cooking a meal
2 Housekeeping and laundry
3 Ability to use the telephone
4 Managing finances
5 Shopping for personal items and food
6 Ability to drive or use public transportation
FIGURE 1 Flow charts of proposed treatment approaches for different presentations of Hodgkin lymphoma. Potential experimental
approaches for which the evidence is not yet mature are shown shaded
42
-
LONGLEY AND JOHNSON
of therapy owing to side effects, but 75% of patients who
completed less than six cycles of AVD remained in CR at
2 years.
20
. Longer followup is needed to ascertain if these re-
sponses are durable and to assess the longterm cardiotoxicity
associated with receiving anthracycline chemotherapy in this pa-
tient cohort which included eight patients >80 years and 9 pa-
tients with a PS of 2.
The combination of N +AVD may prove to be a promising
strategy in fit older patients able to tolerate anthracycline
chemotherapy plus ICI therapy as initial treatment. The Phase II
trial by Ramachandren et al. detailed above included six patients
over 60, five of whom obtained a CR, while one died from toxicity
related to treatment.
16
Data from a larger Phase II trial combining
nivolumab with AVD in older patients with highrisk disease (IPS
score 3) and a positive iPET (DS 4–5) are awaited
(NCT03033914) to assess the activity and safety of this regimen in
a larger cohort of patients.
A chemotherapyfree approach in older patients is an attrac-
tive one in those with poor performance status. A Phase II study
combining nivolumab and BV (N +BV) in elderly patients with a
PS of 0–1 by Yasenchak et al. showed a CR rate of 72% following
16 cycles of treatment that was well tolerated.
22
A subsequent
interim analysis of a similar phase II trial which included a patient
population with poorer PS (0–2) showed a more modest ORR of
61% with CR of 48% following eight cycles of treatment, which
unfortunately did not meet the predefined ORR of 68% and was
therefore terminated early.
23
The frequency of adverse events was
reduced compared to the smaller trial (26% vs. 37% sensory
neuropathy [all grades], respectively) which may be explained by
the difference in the number of cycles received. The strategies to
reduce the number of treatment cycles without compromising
survival should be explored in future, to minimize toxicity and
allow the majority of patients to complete therapy with minimal
dose reductions and treatment interruptions.
The use of comprehensive geriatric assessment (GCA) and quality
oflife (QOL) outcomes in older patients are recommended, but data on
their feasibility are lacking in the prospective clinical trial setting.
Evens et al. used the Cumulative Illness Geriatric (CIRSG)
FIGURE 1 (Continued)
LONGLEY AND JOHNSON
-
43
comorbidity score (https://www.mdcalc.com/cumulativeillnessrat-
ingscalegeriatriccirsg) to stratify patients into high or low groups at
diagnosis (CIRSG score cutoff 10); however, in a multivariable anal-
ysis, this was not significant in predicting PFS, while the loss of
instrumental activities of daily living (IADLs) were predictive of both
OS and PFS, with 2year PFS of 94% versus 25% in patients with no
IADL loss versus IADL loss and 2year OS of 97% versus 67%,
respectively.
20
This assessment is quick and convenient for the busy
oncologist to use in the clinic, to inform the initial treatment decisions,
and help tailor initial therapy to each individual patient's circum-
stances (Table 2).
5
|
CONCLUSIONS
The use of PETdirected therapy in younger patients with advanced
HL has allowed safe deescalation of treatment for those with
responsive disease at iPET, sparing the acute and longterm toxicity
of more intensive chemotherapy regimens and consolidation radio-
therapy. Figure 1summarises the approaches currently in use, and
some of the areas of emerging evidence for new treatments. Further
information regarding highrisk features at diagnosis in the PET
directed era is required, however, and measurement of baseline
PET characteristics may prove a valuable predictive biomarker. The
use of ICI therapy as part of initial therapy, guided by prospective
trials incorporating biomarkers related to the tumor microenviron-
ment is awaited, although the use of PETdirected therapy in this
context is likely to be more complex, with the phenomenon of tumor
flare and delayed response making interpretation harder. The use of
BV and ICIs in older patients appears promising; however, prospec-
tive incorporation of geriatric and QOL assessment into clinical trials
is currently lacking. The use of IADL assessment may provide on-
cologists with a tool to quickly assess patients in the clinic and help
with decisionmaking in this complex patient population.
ACKNOWLEDGMENT
Both authors are supported by grants from Cancer Research UK.
CONFLICT OF INTEREST
Dr. Longley: no conflicts; Dr. Johnson: Research funding from Jans-
sen, Epizyme; Consulting fees from Takeda, BristolMyers Squibb,
Novartis, Genmab, Incyte, Morphosys, and Kymera.
ORCID
Peter W. M. Johnson https://orcid.org/0000-0003-2306-4974
REFERENCES
1. Merli F, Luminari S, Gobbi PG, et al. Longterm results of the
HD2000 trial comparing ABVD versus BEACOPP versus COPP
EBVCAD in untreated patients with advanced Hodgkin lym-
phoma: a study by Fondazione Italiana Linfomi. J Clin Oncol.
2016;34(11):11751181.
2. Borchmann P, Goergen H, Kobe C, et al. PETguided treatment in
patients with advancedstage Hodgkin's lymphoma (HD18): final
results of an openlabel, international, randomised phase 3 trial by
the German Hodgkin study group. Lancet. 2018;390(10114):2790
2802.
3. Johnson P, Federico M, Kirkwood A, et al. Adapted treatment guided
by interim PETCT scan in advanced Hodgkin's lymphoma. N Engl J
Med. 2016;374(25):24192429.
4. Trotman J, Barrington SF. The role of PET in firstline treatment of
Hodgkin lymphoma. Lancet Haematol. 2021;8(1):e67e79.
5. Cottereau AS, Versari A, Loft A, et al. Prognostic value of baseline
metabolic tumor volume in earlystage Hodgkin lymphoma in the
standard arm of the H10 trial. Blood. 2018;131(13):14561463.
6. Pike LC, Kirkwood AA, Patrick P, et al. Can baseline PETCT features
predict outcomes IN advanced Hodgkin lymphoma? A prospective
evaluation OF UK patients in the RATHL trial (CRUK/07/033).
Hematol Oncol. 2017;35(S2):3738.
7. Casasnovas RO, Kanoun S, Tal I, et al. Baseline total metabolic
volume (TMTV) to predict the outcome of patients with advanced
Hodgkin lymphoma (HL) enrolled in the AHL2011 LYSA trial. J Clin
Oncol. 2016;34(15_suppl):7509.
8. Stephens DM, Li H, Schöder H, et al. Fiveyear followup of
SWOG S0816: limitations and values of a PETadapted approach
with stage III/IV Hodgkin lymphoma. Blood. 2019;134(15):1238
1246.
9. Zinzani PL, Broccoli A, Gioia DM, et al. Interim positron emission
tomography responseadapted therapy in advancedstage Hodgkin
lymphoma: final results of the phase II part of the HD0801 study. J
clin oncol. 2016;34(12):13761385.
10. Straus DJ, DługoszDanecka M, Alekseev S, et al. Brentuximab
vedotin with chemotherapy for stage III/IV classical Hodgkin lym-
phoma: 3year update of the ECHELON1 study. Blood.
2020;135(10):735742.
11. Kumar A, Casulo C, Advani RH, et al. A pilot study of brentuximab
vedotin combined with AVD chemotherapy and radiotherapy in
patients with newly diagnosed early stage, unfavorable risk Hodgkin
lymphoma. Blood. 2019;134(Suppl_1):2834.
12. Borchmann P, Plütschow A, Kobe C, et al. PETguided omission of
radiotherapy in earlystage unfavourable Hodgkin lymphoma (GHSG
HD17): a multicentre, openlabel, randomised, phase 3 trial. Lancet
Oncol. 2021;22(2):223234.
13. Casasnovas RO, Bouabdallah R, Brice P, et al. PETadapted treat-
ment for newly diagnosed advanced Hodgkin lymphoma (AHL2011):
a randomised, multicentre, noninferiority, phase 3 study. Lancet
Oncol. 2019;20(2):202215.
14. Engert A, Haverkamp H, Kobe C, et al. Reducedintensity
chemotherapy and PETguided radiotherapy in patients with
advanced stage Hodgkin's lymphoma (HD15 trial): a randomised,
openlabel, phase 3 noninferiority trial. Lancet. 2012;379(9828):
17911799.
15. Reinke S, Bröckelmann PJ, Iaccarino I, et al. Tumor and microenvi-
ronment response but no cytotoxic Tcell activation in classic
Hodgkin lymphoma treated with antiPD1. Blood. 2020;136(25):
28512863.
16. Ramchandren R, DomingoDomènech E, Rueda A, et al. Nivolumab
for newly diagnosed advancedstage classic Hodgkin lymphoma:
safety and efficacy in the phase II CheckMate 205 study. J Clin Oncol.
2019;37(23):19972007.
17. Allen PB, Savas H, Evens AM, et al. Pembrolizumab followed by AVD
in untreated early unfavorable and advanced stage classical Hodgkin
lymphoma. Blood. 2020;137(10):13181326.
18. Cheson BD, Ansell S, Schwartz L, et al. Refinement of the Lugano
Classification lymphoma response criteria in the era of immuno-
modulatory therapy. Blood. 2016;128(21):24892496.
19. Evens AM, Sweetenham JW, Horning SJ. Hodgkin lymphoma in older
patients: an uncommon disease in need of study. Oncology.
2008;22(12):13691379.
44
-
LONGLEY AND JOHNSON
20. Evens AM, Advani RH, Helenowski IB, et al. Multicenter phase II
study of sequential brentuximab vedotin and Doxorubicin, vinblas-
tine, and dacarbazine chemotherapy for older patients with un-
treated classical Hodgkin lymphoma. J Clin Oncol. 2018;36(30):3015
3022.
21. Friedberg JW, ForeroTorres A, Bordoni RE, et al. Frontline bren-
tuximab vedotin in combination with dacarbazine or benda-
mustine in patients aged 60 years with HL. Blood. 2017;130(26):
28292837.
22. Yasenchak CA, Bordoni R, Yazbeck V, et al. Phase 2 study of
frontline brentuximab vedotin plus Nivolumab in patients with
Hodgkin lymphoma aged 60 years. Blood. 2019;134(Suppl_1):237.
23. Cheson BD, Bartlett NL, LaPlant B, et al. Brentuximab vedotin plus
nivolumab as firstline therapy in older or chemotherapyineligible
patients with Hodgkin lymphoma (ACCRU): a multicentre, single
arm, phase 2 trial. Lancet Haematol. 2020;7(11):e808e815.
How to cite this article: Longley J, Johnson PWM.
Personalized medicine for Hodgkin lymphoma: Mitigating
toxicity while preserving cure. Hematological Oncology.
2021;39(S1):3945. https://doi.org/10.1002/hon.2856
LONGLEY AND JOHNSON
-
45
Received: 17 February 2021
DOI: 10.1002/hon.2850
SUPPLEMENT ARTICLE
Cutaneous Tcell lymphomas—An update 2021
Werner Kempf
1,2
|Christina Mitteldorf
3
1
Kempf und Pfaltz Histologische Diagnostik,
Zurich, Switzerland
2
Department of Dermatology, University
Hospital Zurich, Zurich, Switzerland
3
Department of Dermatology, Venereology
and Allergology, University Medical Center
Göttingen, Göttingen, Germany
Correspondence
Werner Kempf, Department of Dermatology,
University Hospital Zurich, Zurich CH8091,
Switzerland.
Email: werner.kempf@uzh.ch
Abstract
Cutaneous Tcell lymphomas (CTCL) represent the majority of primary cutaneous
lymphomas (CL). Mycosis fungoides (MF) and cutaneous CD30+lymphoprolifer-
ative disorders account for 80% of all CTCL. CTCL show overlapping histological
features. Thus clinicalpathological correlation is of importance to achieve final
diagnosis. MF shows a characteristic evolution with patches, plaques, and in a
subset of patients (10%–20%) with tumors. Therapy is stageadapted with skin
directed therapies such as UVlight therapies and corticosteroids in early disease
stage (i.e., patch and limited plaque stage) and systemic therapies (retinoids, inter-
feron, mono chemotherapy, targeted therapy) and/or radiation therapy (local or
total skin beam electron) in advanced stages. Novel therapies include targeted
therapy such as mogamulizumab (antiCCR4) or brentuximab vedotin (antiCD30)
and histone deacetylase inhibitors. Considering the impact of targeted therapies,
biomarkers such as CD30 are not only crucial for the diagnosis and correct classi-
fication of an individual lymphoma case, but also for therapy as they may represent
therapeutic targets. In the recently revised WHO classification 2017 and the
updated WHOEORTC classification for CL 2018, primary cutaneous CD8+acral T
cell lymphoma has been introduced as a new still provisional entity. It displays
characteristic clinical, histological, and phenotypic features and exhibits an excellent
prognosis. Rare, but aggressive CTCL include cutaneous primary cutaneous
aggressive epidermotropic CD8positive Tcell lymphoma and cutaneous gamma/
delta Tcell lymphoma, which present with rapid onset of necrotic or ulcerated
plaques and tumors. As they have a poor prognosis, treatment includes multiagent
chemotherapy and hematopoietic stem cell transplantation.
KEYWORDS
classification, cutaneous, dermatopathology, diagnosis, lymphoma, skin, Tcell, therapy
1
|
INTRODUCTION
The group of primary cutaneous lymphomas (CL) are the second most
common group of extranodal lymphomas. Among CL, cutaneous T
cell lymphomas (CTCL) represent the majority accounting for
approximately 65%–75% of all CL.
1
All other CTCL forms are rare
with each entity accounting for less than 1% of all CTCL. Primary CL
present in the skin without extracutaneous disease at the time of
diagnosis and often remain limited to the skin over long periods of
disease evolution. Progression with extracutaneous spread occurs
usually in advanced late stages.
CTCL exhibit a wide spectrum of clinical, histological, immuno-
phenotypic features and genetic alterations. As these histological and
phenotypic features overlap within the various forms of CTCL, cor-
relation of the clinical features with the histological and immuno-
phenotypic findings is an essential element of the diagnostic workup
46
-
Hematological Oncology. 2021;39(S1):4651. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
and implies that clinical images and dermatologic examination are
indispensable to achieve the final diagnosis. The prognosis and
treatment significantly differ among the entities within the CTCL
group and also from systemic lymphomas with similar histological
features.
Recent developments such as targeted therapy have expanded
the therapeutic spectrum in CTCL and contribute to a significant
change in the prognosis, particularly in advanced forms of several
CTCL forms. The therapeutic response to a distinct targeted therapy,
however, varies among different CTCL entities. The exact classifica-
tion of CTCL entities is important to compare results of clinical trials
and to assess the value of targeted therapy. CTCL without further
specification should not be used as a diagnostic term in therapeutic
trials as this designation does not allow to identify differences in the
responses to therapy among the distinct CTCL forms. The classifi-
cation and terminology of CL should follow the current revised WHO
classification 2017 (Table 1) and the updated WHOEORTC classifi-
cation for CL 2018.
1,2
An example for the efficacy of targeted therapy in CTCL is
brentuximab vedotin (BV), which targets CD30 expressed on
tumor cells. It is not only a new and efficient treatment for
relapsed or therapy refractory CD30 lymphoproliferative disor-
ders (LPDs) such as multifocal primary cutaneous anaplastic large
cell lymphoma (PCALCL) but also expanded the therapeutic
strategies for advanced mycosis fungoides (MF) and Sézary
syndrome.
3
Moreover, BV was even effective in some patients in
whom rare and aggressive CTCL forms such as cutaneous
gamma/delta Tcell lymphoma showed exceptional expression of
CD30.
Considering the impact of targeted therapies, biomarkers such as
CD30 are not only crucial for the diagnosis and correct classification
of an individual lymphoma case, but also for therapy as they may
represent therapeutic targets. In consequence, the detection of
biomarker expression by immunohistochemistry or other in situ
techniques became an important aspect in the diagnostic workup of
CTCL. Interobserver variability and differences in the technical as-
pects, however, are critical factors in the assessment of biomarkers.
Thus, detection methods and expression level need to be standard-
ized especially in situations, in which the expression level of a target
is decisive for the selection of therapy. Digital image analysis tools
are currently developed for the automatized evaluation of immuno-
histochemical slides with the aim to reduce the interinstitutional and
interrater variability. Nevertheless, the results should always be
interpreted in a synoptic way taking into account the context of
clinical and histopathological context, especially since CD30 expres-
sion is found in a broad range of CTCL.
4
2
|
CTCL ENTITIES
MF is the most common form of CTCL and accounts for nearly
50% of all primary CL.
1
MF shows a characteristic disease evo-
lution with erythematous patches (patch stage) which may evolve
into more infiltrated plaques (plaque stage). The patches and
TABLE 1Spectrum of primary
cutaneous Tcell lymphomas according
to revised WHO classification of
hematopoietic tumors (2017) and the
updated WHOEORTC classification for
primary cutaneous lymphomas (2018)
Mycosis fungoides (MF)
MF variants
Folliculotropic MF
Granulomatous slack skin
Pagetoid reticulosis
Sézary syndrome
Adult Tcell leukemia/lymphoma
Primary cutaneous CD30positive lymphoproliferative disorders
Lymphomatoid papulosis
Primary cutaneous anaplastic large cell lymphoma
Subcutaneous panniculitislike Tcell lymphoma
Primary cutaneous gamma/delta Tcell lymphoma
Primary cutaneous aggressive epidermotropic CD8positive Tcell lymphoma (provisional)
Primary cutaneous CD4+small/mediumsized pleomorphic Tcell lymphoproliferative disorder
(provisional)
Primary cutaneous acral CD8+Tcell lymphoma (provisional)
Primary cutaneous peripheral Tcell lymphoma, not otherwise specified (NOS)
Extranodal NK/Tcell lymphoma, nasal type
Hydroa vacciniformelike lymphoproliferative disorder
Abbreviation: MF, mycosis fungoides.
KEMPF AND MITTELDORF
-
47
plaques usually persist over long time, that is, months or years. In
a subset of patients, large and often ulcerated tumors (tumor
stage) develop.
As the histological findings may be subtle especially in patch
stage of MF, distinction from benign inflammatory skin diseases
such as chronic eczema may not always be possible at this disease
stage and require repeated biopsies from different lesions. This may
also explain the delay of several months to years in the diagnosis of
early MF.
Enlarged lymph nodes with clinical and/or radiologic suspicion
for lymphoma involvement should be either biopsied. Bone marrow
biopsy is only indicated in advanced disease stage or if circulating
atypical lymphocytes in the peripheral blood are present.
Histologic workup is not only important to identify MF and to
distinguish MF from inflammatory skin diseases, but it also allows to
identify variants and subtypes of MF which may differ in their course
from classic MF. Folliculotropic MF (FMF) is a distinct variant of MF
accounting for approximately 10% of all MF cases and characterized
by folliculotropic infiltrates, that is, exocytosis of tumor cells into the
hair follicle epithelia resulting in alopecic patches and plaques.
5
Recently, two prognostic subsets of FMF could be identified. The
early form of FMF manifests with follicular papules, acneiform lesions
with comedones and cysts and histologically with subtle lymphocytic
infiltrates. The advanced form of FMF is characterized by more
extensively infiltrated thicker alopecic plaques and deep, dense
lymphocytic infiltrates. The course and prognosis of advanced form is
similar to tumor stage of MF, whereas the early form displays the
same indolent course as the classic form of MF (patch stage).
The prognosis in the patch and limited plaque stage is favorable
with 5and 10year survival rates over 90% reflecting the slowly
progressive and indolent course of the disease in most patients. In
stage IB the 10yearsurvival rate is 80% but drops to 40% in stage
IIB.
6
whereas the prognosis in the patch and limited plaque stage is
favorable. MF in tumor stage is aggressive and associated with the
risk for extracutaneous spread which often first involves lymph
nodes. During disease progression, visceral or other organs may
become involved and sepsis originating from bacterial colonization of
ulcerated tumors. Remarkably, the majority of patients (80%) will not
progress to tumor stage. Thus, it is important to emphasize this point
in counseling MF patients as they will get often terrified when
looking to images of advanced MF seen during their Internet
searches to retrieve more information about the disease and the
treatment modalities.
Treatment of MF is stageadapted which is also reflected in in-
ternational and national guidelines (Table 2).
Skindirected therapies (UV light, topical corticosteroids, nitro-
gen mustard) are the main strategies for early stage of MF and sys-
temic therapies (retinoids, chemotherapy, targeted therapy) for
advanced disease (extensive plaque and tumor stage).
7,8
It should be noted that in selected patients with stage IA (limited
patch stage) a watchandwait strategy may be justified. In early MF
(patch and limited plaque stage), skindirected therapies comprise
treatment by UVlight (UVB narrow band, psoralen UVA [PUVA]),
topical corticosteroids (CS) and the newly introduced topical mech-
lorethamine (nitrogen mustard) representing a topically applied
cytotoxic agent. Topical CS and UVlight approaches result in com-
plete clinical response in up to 60% and 80%–90% of the patients,
respectively. These therapies should be intermittently applied to
prevent or reduce the longterm complications of UVlight treatment
(epithelial skin cancers) and toxicity of potent topical corticosteroids
(skin atrophy and fragility). Topical mechlorethamine (nitrogen
mustard) represents a novel therapy for MF in patch and limited
plaque stage with response rates up to 60% of the patients and
represents an alternative, but more costintense treatment for pa-
tients not responding to other skindirected therapies. Irritative and
allergic dermatitis are seen as adverse effects.
In patients with relapses or therapy refractory course and
development of plaques, skindirected treatment is combined with
systemic treatment including retinoids and retinoid analogues (aci-
tretin; bexarotene) or lowdose methotrexate. Interferonalpha (IFN),
which was an effective alternative to retinoids and reached response
TABLE 2Treatment of MF
Topical therapies:
Corticosteroids ointments
Bexarotene gel (US)
Chlormethine gel
UVlight therapies:
UVB narrow band (if patches only)
PUVA (alone or in combination with other therapies)
Systemic therapies:
Interferon alpha
a
Oral bexarotene
Mogamulizumab (antiCCR4)
Brentuximab vedotin (antiCD30 coupled with monomethyl
auristatin E)
Histon deacetylase inhibitors
Chemotherapy:
Methotrexate (low dose) or pralatrexate (US)
Gemcitabine (lowdose)
Pegylated liposomal doxorubicin
CHOP (extracutaneous spread)
Radiotherapy:
Local radiotherapy (solitary or few tumors)
Total skin electron beam (generalized thick plaques and tumors).
Extracorporeal photopheresis (erythrodermic MF)
Allogeneic hematopoietic stem cell transplantation
Abbreviation: MF, mycosis fungoides.
a
Production of nonpegylated INF alpha 2a stopped, alternatively
pegylated interferon alpha (limitation: not approved for MF).
48
-
KEMPF AND MITTELDORF
rates of more than 50% in combination with PUVA, is not any longer
produced. The nonpegylated INF alpha is not any longer available,
but pegylated INF alpha might be an alternative. Similar response
rates can be achieved by the combination of PUVA and retinoids
(acitretin or bexarotene).
As an alternative to systemic retinoids, topical retinoids such as
alitretinoin, bexarotene, and tazarotene can be used. As in mechlor-
ethamine, irritation is a common adverse effect. In patients with
single tumors in otherwise patch and plaque stage, local radiotherapy
for tumors is effective.
Recently, new treatment options became available such as moga-
mulizumab, a humanized antibody against CCR4 (CCchemokine
receptor 4) as targeted therapy which leads to a prolonged
progressionfree survival. Histone deacetylase inhibitors (HDACi),
such as vorinostat, romidepsin, and resminostat, interact with various
pathways involved in tumor cell growth. Clinical response to HDACi
therapy is observed in approximately onethird of patients. Resmino-
stat is tested in clinical trials as a maintenance therapy and seems to
enhance response to other therapies.
In more advanced disease (thick plaques, tumor stage), mono-
chemotherapy with gemcitabine or pegylated liposomal doxorubicin
are effective. CHOPbased chemotherapy regimens should be
restricted to patients with extracutaneous spread and usually result
in only shortlived response. Moreover, they seem to increase the
risk for sepsis. Neoplastic cells in MF tumors may express CD30
which renders them to candidates for a targeted therapy with BV, an
antiCD30based antibody conjugated to monomethyl auristatin E as
a cytostatic drug. In cases with more than 5% of the tumor cells
expressing CD30, clinical response to BV was observed and could
induce regression of even large, ulcerated tumors. Alternatively, total
body skin electron beam irradiation is effective for patients with
multiple MF tumors. This treatment modality is, however, only
available in certain centers. Allogeneic hematopoietic stem cell
transplantation is a potentially curative treatment option and is
effective in up to 60% of patients with advanced MF and Sézary
syndrome but is limited by the high relapse rate and the high mor-
tality associated with this therapy.
9
Primary refractory disease,
relapse, or progression in patients that had received three systemic
treatments prior to transplant were identified as independent
adverse prognostic factors. Pretreatment with total skin electron
beam therapy may be beneficial before stem cell transplantation.
7
Bacterial toxins, especially those of Staphylococcus aureus, are
able to activate STAT3 and stimulate proliferation of tumor cells, and
suppress the activity of tumorinfiltrating CD8+cells against the
tumor cells. Antibiotic treatment resulted in clinical improvement in
MF as well as in Sézary syndrome.
10
Sézary syndrome (SES) is an aggressive form of CTCL, which is
derived from central memory Tcells. Clinically, SES manifests with
generalized skin involvement with erythroderma accompanied by
intense pruritus, enlarged lymph nodes (with or without specific
involvement by the neoplastic cells), and a leukemic spread of tumor
clone. Diagnosis requires demonstration of the same tumor clone in
the skin lesions and the peripheral blood as well as additional he-
matologic findings (absolute Sézary cell count of >1000 cells/µl or an
expanded CD4+Tcell population leading to a CD4/CD8 ratio 10,
CD4+/CD7cells 40% or CD4+/CD26cells 30%). Histological
diagnosis may be challenging as nonspecific findings are common.
Immunohistochemical (bio)markers such as PD1, TOX, CD7,
KIRDL2, and the proliferation rate were shown to be useful markers
in the histopathological diagnosis of SES and the distinction from
benign inflammatory erythrodermic skin diseases such as atopic
dermatitis or drug eruption.
11
Other Tcell nonHodgkin lymphomas
present with a leukemic phase and may involve the skin such as Tcell
large granular lymphocytosis (TLGL), Tcell prolymphocytic leukemia
(TPLL), and adult Tcell lymphoma/leukemia (ATLL). The tumor cells
in TLGL are large with abundant cytoplasm and azurophilic granula.
In contrast to SES, the course of TLGL is usually indolent. Expression
of TCL1 detected by immunohistochemistry or flow cytometry is
helpful in the diagnosis of TPLL. ATLL is linked to human Tcell
lymphotropic virus 1 which differs from SES, TLGL, and TPLL.
Treatment of SES includes extracorporeal photopheresis (ECP) with a
wide range of response rates up to 80% and complete response rates
up to 25%.
7
ECP is often combined with systemic therapies used in
MF, for example, bexarotene. Alemtuzumab and chemotherapy
(gemcitabine, pegylated liposomal doxorubicin) are used as second
line treatment. Recently mogamulizumab was reported as an effec-
tive treatment in SES.
Primary cutaneous CD30positive lymphoproliferative disorders
(CD30+LPD) are the second most common group of CTCL (25% of all
CTCL) and include PCALCL, lymphomatoid papulosis (LYP) and
borderline lesions.
4,12
CD30+LPD exhibit an indolent course with a
good prognosis despite recurrences are frequent. PCALCL manifests
with a solitary or grouped, rapidly growing, and often ulcerated large
tumor(s). Histologically, dense cohesive infiltrates of predominantly
large pleomorphic or anaplastic tumor cells which express CD30 and
show a variable loss of Tcell markers. PCALCL does not have a
t(2;5) and thus is negative for anaplastic lymphoma kinase (ALK).
Rearrangements of the DUSP22IRF4 locus are found in approxi-
mately 2530% of PCALCL and rearrangements of TP63 are very
rare in PCALCL. In contrast to systemic ALCL, DUSP22IRF4 rear-
rangemnt does not have a prognostic impact in PCALCL. CD30
expression is not restricted to CD30+LPD, in which the vast majority
of tumor cells express this marker. Therefore, other CTCL such as
MF, particular in tumor stage, need to be distinguished from PC
ALCL and LYP by correlation of the histological findings with clinical
presentation to achieve correct diagnosis within the group of CTCL.
Despite relapses are common, PCALCL has a favorable prognosis
with 5year survival rate of 90%. Patients with extensive limb disease
and multifocal PCALCL are at risk to develop extracutaneous spread
and show impaired prognosis.
Surgical excision or radiation therapy is the firstline treatment
for solitary or localized PCALCL. For patients with multifocal lesions,
lowdose methotrexate (MTX; up to 25 mg/kg per week) or local
radiotherapy in case of only a few lesions is recommended. BV is an
KEMPF AND MITTELDORF
-
49
effective treatment with high response rates in patients who do not
respond to MTX or develop extracutaneous spread to locoregional
lymph nodes.
3,8
Multiagent chemotherapy is indicated only in pa-
tients with extracutaneous spread to visceral organs or CNS or lack
of response to BV.
LYP presents with grouped or generalized papules and small
nodules which show the characteristic spontaneous regression of the
individual lesions within a few weeks. Histology shows a broad
spectrum with five different subtypes (referred to as type A–E) and a
subtype with chromosomal rearrangements involving the DUSPIRF4
locus on 6p25.3 listed in the revised WHO and updated WHO
EORTC classification.
1,2,4
Some of these histological types pose
diagnostic difficulties as they simulate aggressive CTCL (e.g., type D
and E). Thus clinicalpathological correlation is crucial for the diag-
nosis as LYP displays overlapping histological features with other
CTCL forms. LYP carries an excellent prognosis with 5 and 10year
survival rate of almost 100%. Nevertheless, LYP patients are at risk
to develop a second lymphoma, especially MF and cutaneous or nodal
ALCL which occurs in 15% of the patients.
In regard to the excellent prognosis, a watchandwait strategy
for LYP can be justified. UVlight (UVB narrow band or PUVA) and
lowdose methotrexate (5–20 mg per week) are the most common
therapies, but relapses after withdrawal of treatment are common.
13
Thus, the effects of these therapies need to be balanced against long
term adverse effects. For patients with disseminated lesions and lack
of response to MTX or UVlight therapies, BV represents a new
therapeutic strategy and has been used with lower dose than in other
CD30+Tcell lymphomas. For larger lesions (i.e., 2–3 cm in diameter)
which do not regress within 3 months, surgical excision or local
radiotherapy can be applied.
Subcutaneous panniculitislike Tcell lymphoma (SPTCL) is a cyto-
toxic Tcell lymphoma defined by infiltrates of mostly CD8+pleo-
morphic lymphocytes in the lobuli of subcutaneous fat tissue.
1,2
The
neoplastic cells express TCR alpha/beta which distinguishes them
from expression of TCR gamma/delta by the tumor cells in the sub-
cutaneous form of gamma/delta Tcell lymphoma. SPTCL may affect
children as well as adults and presents with subcutaneous plaques or
nodules which involve preferentially the extremities. Regression of
lesions results in focal lipatrophy. In addition to the skin lesions,
systemic symptoms including fever, fatigue and weight loss, cytope-
nia, and elevated liver enzymes can be present. Some patients suffer
from systemic lupus erythematosus. There are overlapping features
between lupus panniculitis to SPTCL.
In most patients, SPTCL shows an indolent course with a favor-
able prognosis (5yearSR: 80%).
1
In 15% of the patients, SPCTL is
complicated by hemophagocytic lymphohistiocytosis (HLH), which is
linked to an aggressive course with high mortality. Recently, germline
mutations in HAVCR2 were found in 60% of SPTCL which result in
loss of function or missense variants of Tcell immunoglobulin mucin
3 as a modulator of immune responses expressed on subgroups of T
and innate immune cells.
14
This results in persistent immune
activation and increased production of inflammatory cytokines which
promote HLH and SPTCL. The hyperactive immune reaction in SPTCL
may explain the fact that immunosuppressive therapies are effective
in these patients. Immunosuppressive drugs such as steroids, MTX,
and cyclosporine have replaced chemotherapy as firstline therapy
for SPTCL, independent of the presence of HLH.
8
CD8+acral Tcell lymphoma (CD8+ATCL) is a new provisional
CTCL entity introduced for the first time in the revised WHO clas-
sification 2017 and updated WHOEORTC classification 2018.
1,2
This
lymphoproliferation manifests with a solitary or rarely bilateral
nodule(s) at acral sites, that is, ears, face, and feet. Histology is
characteristic and shows a dense dermal infiltrate of small to me-
diumsized atypical lymphocytes with moderate nuclear atypia.
15
There is no epidermotropism of neoplastic cells in CD8+ATCL which
differs from the epidermotropic infiltrates in CD8+MF and CD8+
AECTCL. The tumor cells in CD8+ACTL exhibit a characteristic
phenotype with expression of CD3+, CD8+, and TIA1+in the
absence of other cytotoxic markers such as granzyme B and perforin.
In the CD68 stain, a characteristic perinuclear dotlike pattern is
observed. The course is indolent with an excellent prognosis. Staging
is not necessary in cases with typical clinical presentation and his-
tology. CD8+ACTL can be treated with surgical excision or
radiotherapy.
3
|
CONCLUSIONS
CTCL are characterized by distinct clinical features but show signif-
icant overlap in histological and phenotypic features. Thus clinical
pathologic correlation is of outmost importance in the diagnostic
workup of CTCL. MF and cutaneous CD30+LPD represent the most
common form of CTCL. The therapeutic approach for MF depends on
disease stage and evolution. CD30 LPD exhibit a favorable prognosis.
Therefore, unnecessary aggressive treatment should be avoided.
Targeted therapy are novel and effective treatment modalities for
advanced MF and in patients with therapy refractory CD30 LPD.
CONFLICT OF INTEREST
The authors have no conflict of interest to declare.
PEER REVIEW
The peer review history for this article is available at https://publons.
com/publon/10.1002/hon.2850
ORCID
Werner Kempf https://orcid.org/0000-0002-6552-8629
REFERENCES
1. Willemze R, Cerroni L, Kempf W, et al. The 2018 update of the
WHOEORTC classification for primary cutaneous lymphomas.
Blood. 2019;133(16):17031714.
50
-
KEMPF AND MITTELDORF
2. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tu-
mours of Haematopoietic and Lymphoid Tissues. Lyon: IARC; 2017.
3. Engelina S, Saggu M, Yoo J, et al. Brentuximab a novel antibody
therapy: realworld use confirms efficacy and tolerability for CD30
positive cutaneous lymphoma. Br J Dermatol. 2020;182(3):799800.
4. Kempf W, Kerl K, Mitteldorf C. Cutaneous CD30positive Tcell
lymphoproliferative disordersclinical and histopathologic features,
differential diagnosis, and treatment. Semin Cutan Med Surg.
2018;37(1):2429.
5. Mitteldorf C, Stadler R, Sander CA, Kempf W. Folliculotropic
mycosis fungoides. J Dtsch Dermatol Ges. 2018;16(5):543557.
6. Mourad A, Gniadecki R. Overall survival in mycosis fungoides: a
systematic review and metaanalysis. J Invest Dermatol.
2020;140(2):495497.
7. Trautinger F, Eder J, Assaf C, et al. European Organisation for
Research and Treatment of Cancer consensus recommendations for
the treatment of mycosis fungoides/Sezary syndrome update 2017.
Eur J Cancer. 2017;77:5774.
8. Willemze R, Hodak E, Zinzani PL, Specht L, Ladetto M, Committee
EG. Primary cutaneous lymphomas: ESMO Clinical Practice Guide-
lines for diagnosis, treatment and followup. Ann Oncol.
2018;29(Suppl 4):iv30iv40.
9. Iqbal M, Reljic T, Ayala E, et al. Efficacy of allogeneic hematopoietic
cell transplantation in cutaneous T cell lymphoma: results of a sys-
tematic review and metaanalysis. Biol Blood Marrow Transpl.
2020;26(1):7682.
10. Lindahl LM, WillerslevOlsen A, Gjerdrum LMR, et al. Antibiotics
inhibit tumor and disease activity in cutaneous Tcell lymphoma.
Blood. 2019;134(13):10721083.
11. Klemke CD, Booken N, Weiss C, et al. Histopathological and
immunophenotypical criteria for the diagnosis of Sezary syndrome in
differentiation from other erythrodermic skin diseases: a European
Organisation for Research and Treatment of Cancer (EORTC)
Cutaneous Lymphoma Task Force Study of 97 cases. Br J Dermatol.
2015;173(1):93105.
12. PrietoTorres L, RodriguezPinilla SM, Onaindia A, Ara M, Requena
L, Piris MA. CD30positive primary cutaneous lymphoproliferative
disorders: molecular alterations and targeted therapies. Haemato-
logica. 2019;104(2):226235.
13. Kempf W, Pfaltz K, Vermeer MH, et al. EORTC, ISCL, and USCLC
consensus recommendations for the treatment of primary cuta-
neous CD30positive lymphoproliferative disorders: lymphomatoid
papulosis and primary cutaneous anaplastic largecell lymphoma.
Blood. 2011;118(15):40244035.
14. Gayden T, Sepulveda FE, KhuongQuang DA, et al. Germline
HAVCR2 mutations altering TIM3 characterize subcutaneous pan-
niculitislike T cell lymphomas with hemophagocytic lymphohistio-
cytic syndrome. Nat Genet. 2018;50(12):16501657.
15. Petrella T, Maubec E, CornilletLefebvre P, et al. Indolent CD8
positive lymphoid proliferation of the ear: a distinct primary cuta-
neous Tcell lymphoma? Am J Surg Pathol. 2007;31(12):18871892.
How to cite this article: Kempf W, Mitteldorf C. Cutaneous T
cell lymphomas—An update 2021. Hematological Oncology.
2021;39(S1):4651. https://doi.org/10.1002/hon.2850
KEMPF AND MITTELDORF
-
51
Received: 18 February 2021
DOI: 10.1002/hon.2846
SUPPLEMENT ARTICLE
What's new in peripheral Tcell lymphomas
Stefano Luminari
1,2
|Tetiana Skrypets
1,3
1
Surgical, Medical and Dental Morphological
Sciences Related to Transplant, Oncology and
Regenerative Medicine, University of Modena
and Reggio Emilia, Reggio Emilia, Italy
2
Hematology Unit, Azienda USL IRCCS of
Reggio Emilia, Reggio Emilia, Italy
3
PhD Programm in Clinical and Experimental
Medicine, University of Modena and Reggio
Emilia, Italy
Correspondence
Stefano Luminari, Medical and Dental
Department of Morphological Sciences
Related to Transplant, Oncology and
Regenerative Medicine, University of Modena,
Viale risorgimento 80, 42123 Reggio Emilia,
Italy.
Email: sluminari@unimore.it
Abstract
Peripheral Tcell lymphomas (PTCLs) are a rare, heterogeneous group of hemato-
logical malignancies with extremely poor prognosis for almost all subtypes. The
diverse clinicopathological features of PTCLs make accurate diagnosis, prognosis,
and choice of optimal treatment strategies difficult. Moreover, the best therapeutic
algorithms are still under debate due to the extrapolated approaches developed for
Bcell lymphomas and to the absence of few treatment protocol specifically
developed for PTCLs. Some advances have been made with CD30 monoclonal
antibody, mainly for anaplastic largecell lymphomas, with improvements in
progressionfree survival and overall survival. Several new drugs are under evalu-
ation in clinical trials, although not all the results are as encouraging as expected. In
this review, we briefly present the most updated information on diagnosis, prog-
nostication, and treatment strategies in PTCLs.
KEYWORDS
chemotherapy, novel agents, peripheral Tcell lymphoma, prognostic factors
1
|
INTRODUCTION
Peripheral Tcell lymphomas (PTCLs) comprise a heterogeneous
subgroup of rare hematological malignancies originating from post-
thymic lymphocytes. From different available data, PTCLs account
for approximately 5%–15% of all lymphomas in Western countries,
with an incidence of 0.5–2 per 100,000 people per year.
1,2
Four main
clinical subtypes have been identified: nodal, leukemic, disseminated,
and cutaneous PTCLs. From a pathologic point of view, the most
recent edition of the World Health Organization (WHO) Classifica-
tion of lymphomas identifies around 30 subtypes of PTCLs, now also
defined as mature Tcell lymphomas (MTCLs), some of which are
extremely rare. The most common MTCL subtypes are PTCL not
otherwise specified (NOS), angioimmunoblastic Tcell lymphoma
(AITL), anaplastic largecell lymphoma (ALCL), and natural killer (NK)/
Tcell lymphoma (NKTCL).
1,3,4
With the exception of cutaneous Tcell
lymphomas (CTCL), which are usually characterized by an indolent
course, and of anaplastic lymphoma kinase (ALK) positive ALCL, all
other MTCLs are associated with an aggressive course and poor
outcomes, with 5year survival rates across subtypes of 30%.
3
Improvement in the approach to MTCL is proceeding slowly, with
advances in recent years seen in improving the classification of
MTCL, patient management and prognostication, and treatment.
In the 2016 revision of the WHO of mature Tcell neoplasms,
nodal lymphomas of T follicular helper (TFH) cell origin were
introduced. AITL is the most studied TFH subtype, but an addi-
tional 40% of cases of PTCLNOS have been shown to share some
of the clinical and pathological features of the TFH phenotype,
which requires the expression of at least two of three TFHrelated
antigens, including PD1, CD10, BCL6, CXCL13, ICOS, SAP, and
CCR5. Recurrent genetic abnormalities associated with TFH
phenotype include mutations of in epigenetic modifiers (TET2,
IDH2, DNMT3A), RHOA, and Tcell receptor associated genes
(PLCG1, CD28, VAV1, FYN).
3,5
Among the nonTFH PTCLNOS,
gene expression profiling and microRNA profiling studies have
delineated two additional subgroups: those with an increased
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, pro-
vided the original work is properly cited.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
52
-
Hematological Oncology. 2021;39(S1):5260. wileyonlinelibrary.com/journal/hon
expression of GATA3 and those with an increased expression of
TBX21.
3,5
The GATA3 group is associated with poor outcomes and
has more loss or mutation of tumor suppressor genes including
TP53, PTEN, PRDM1, and CDKN2A/B, and gains in STAT3, REL,
and MYC oncogenes.
6
The TBX21 group is enriched with alter-
ations of genes involved in DNA methylation.
7
These three groups
are added to other previously characterized subtypes with specific
cell of origin, which include ALCL, adult Tcell leukemia/lymphoma,
and gammadelta PTCL. The better characterization of the cell of
origin in PTCL is advantageous in the classification of these lym-
phomas as it reduces the undefined basket of PTCLNOS and
provides a strong rationale for determining the most effective
therapies in these lymphomas. Immunomodulatory agents and
epigenetic modifiers are more suitable for TFH subtypes, while
phosphatidylinositol 3kinase (PI3K) inhibitors, hypomethylating
agents, and Janus kinase/signal transducer and activator of tran-
scription (JAK/STAT) inhibitors might find a better role in GATA3
or TBX21 subtypes.
8
1.1
|
Prognosis and staging
In the last two decades, many studies have been conducted to
identify and validate clinical and biological factors that can be used
to predict the heterogenous outcome of PTCL patients. Several of
these studies have confirmed that a poor Eastern Cooperative
Oncology Group Scale of Performance Status score, extranodal
involvement, advanced disease, bulky, Ki67, and a high lactate de-
hydrogenase rate correlate with shorter overall survival (OS).
9–11
The International Prognostic Index (IPI) was formally validated in
PTCL
1,12
but the lack of clearly defined risk groups prompted the
search for PTCLspecific prognostic indexes.
9,11
The Fondazione
Italiana Linfomi defined the first Prognostic Index for PTCLNOS
(PIT), which stratifies patients into four distinct groups; the PIT
showed an independent correlation with OS. Subsequently, PIT
was updated to modified PIT by replacing bone marrow involve-
ment with Ki67 rate expression.
10
A new model, the Tcell
score, has recently been defined by using the prospectively
collected data registered in the Tcell Project.
13
More recently,
novel prognostic indexes have been identified and validated for
specific PTCL subtypes, including enteropathyassociated Tcell
lymphoma (EATCL) and nasaltype extranodal NKTCL. As shown
in Table 1, all available prognostic indexes share the same structure
of categorical scores based on simple laboratory and clinical fea-
tures. Although all of these indexes have been formally validated,
the accuracy of prognostication in PTCL remains suboptimal; thus,
more prognostic studies that take into account novel biomarkers
and novel prognostic features are warranted. Among recent ad-
vances, a better definition of response by means of
18
F
fluorodeoxyglucose positron emission tomography (FDGPET) may
play an important role in PTCL management and decision making.
14
PTCLs are listed among FDGavid diseases, and several studies
have already confirmed the role of metabolic tumor volume and of
interim and endoftreatment FDGPET to predict outcomes.
15,16
Although promising, data regarding the role of metabolic response
in PTCLs are very preliminary and thus need to be confirmed by
larger studies.
1.2
|
Treatment
The optimal management of patients with PTCL, which is disputed, is
in any case limited to few options, all with unsatisfactory efficacy.
None of the currently available recommendations are based on high
quality evidence, and few welldesigned randomized clinical trials
(RCTs) have been conducted to support therapeutic choices. The
currently recommended treatment strategy for PTCLs derives mostly
from Bcell lymphoma treatment strategies, with the recommended
use of an aggressive approach with anthracyclinebased poly-
chemotherapy (i.e., CHOP or CHOEP) and with autologous stem cell
transplant (autoSCT) to consolidate response to firstline therapy or
to manage relapsed patients.
17
Regarding the role of anthracyclines in PTCL, while their role is
still debated, anthracycline void regimens have so far failed
to demonstrate their superiority to CHOP.
17
Based on a recent
metaanalysis, the 5year OS achieved with this approach was
36.6%.
18
Several attempts have been made to improve the poor results
achieved with CHOP. These include the addition of novel agents
and the intensification of therapy. Some clinical studies on etopo-
side intensification of standard CHOP have shown conflicting re-
sults. However, the addition of etoposide has shown better
progressionfree survival (PFS), especially in patients with ALCL,
in those with favorable risk factors, and in patients age 60
years.
12,19–21
The results of three randomized trials that evaluated the efficacy
of adding a novel agent to the CHOP backbone are available.
One prospective trial combined alemtuzumab, an antiCD52
monoclonal antibody, with CHOP; it failed to show improved out-
comes compared to chemotherapy alone.
22
Another randomized trial
compared standard CHOP with CHP (cyclophosphamide, vincristine,
prednisone) combined with the antiCD30 antibodydrug conjugate
brentuximab vedotin (BVCHP; ECHELON2 trial).
23
This trial, which
enrolled 452 treatment naïve CD30positive PTCLs, demonstrated
improved PFS and OS rates for the BVCHP combination.
23
Most of
the patients included in this trial (about 75%) had ALCL, with clearly
positive results for this subtype. However, the scientific community
was left without any clear demonstration of the efficacy of BVCHP
in nonALCL CD30positive subtypes.
A third trial compared CHOP with CHOP +the histone deace-
tylase inhibitor romidepsin, showing promising singleagent activity
in relapsed refractory patients (ROCHOP trial).
24
The ROCHOP RCT
conducted by the LYSA group enrolled 421 patients with PTCL who
were not planned to receive autoSCT or allogeneic SCT (alloSCT).
The median PFS (mPFS) for patients in the experimental arm was 12
months (9–25.8), without significant difference compared to the
LUMINARI AND SKRYPETS
-
53
reference arm (mPFS 10.2 months [7.4–13.2]; hazard ratio: 0.81; 95%
confidence interval: 0.63–1.04). Although this study was not able to
confirm the initial hypothesis of the superiority of RoCHOP, the
subgroup analysis seems to suggest that the novel combination acts
differently in different PTCL subtypes, with relatively higher activity
observed for AITLs.
24
In summary, even if associated with unsatisfactory results, CHOP
chemotherapy should still be considered as the reference therapy for
most PTCL subtypes with the main exception of ALCL for which BV
CHP is the preferred recommended option and of NKTCL. The use of
CHOEP is supported by low quality of evidence but can be consid-
ered as a reasonable option in young and fit subjects with nonALCL
PTCLs.
The use of highdose chemotherapy followed by autoSCT in
first complete remission (CR1) is recommended by most of the
available guidelines
17,25
(Table 2). Several groups have reported
that achieving CR before autoSCT is a significant independent
predictor of improved survival in patients with PTCL receiving
upfront autoSCT.
26–28
However, there have been no RCTs specif-
ically designed to evaluate upfront autoSCT in comparison
with observation in CR1 for PTCL.
29–31
Several retrospective
studies and prospective singlearm phase II trials have reported
encouraging results with this approach (Table 3). The largest
prospective phase II study, published by the Nordic Group (NLGT
01), included 160 patients with PTCLs; 72% of patients underwent
autoSCT in first remission after six courses of CHOEP chemo-
therapy.
32
All nodal PTCL subtypes were included, with the
exception of A.
1.2.1
|
ALKpositive ALCLs
One hundred thirty patients achieved CR (63%) or partial response
(PR; 37%), and 115 (88.5%) underwent ASCT. Overall, the 5year
OS and PFS for the intentiontotreat population were 51% and
44%, respectively. Considering subtype distribution, better out-
comes were observed for ALKnegative ALCL than for other sub-
types.
32
In a second study by Reimer et al.,
33
83 patients with
PTCL were enrolled, with the exclusion of CTCL and of ALK
positive ALCL. Fiftynine patients (71%) completed stem cell
mobilization after CR (66%) or PR (34%) and 55 underwent
autoSCT. The 3year OS was 48%.
TABLE 1Prognostic models in PTCL
Variable
IPI PIT IPTCLP mPIT TCS AITL PINK EPI
International
NHL Prognostic
Factors Project (1993)
Gallamini
et al.
(2004)
Vose
et al.
(2008)
Went
et al.
(2006)
Federico
et al.
(2018)
Hong
et al.
(2018)
Kim
et al.
(2016)
de Baaij
et al.
(2015)
Age >60 X X X X X X
ECOG 1X X X X x X
LDH (abn. values) X X X X X
Stage III–IV X x X X X
ENS >2X X X
BM+X
Plt <150 K/mmc X X
SAlb x
Neutrophils x
Ki67 80% X
Anemia (M <13,
F<11g/dl)
X
Serum IgA
(>400 mg/dl)
X
B symptoms X X
Regional
lymph nodes
X
PTCL subset All PTCLNOS PTCLNOS PTCLNOS PTCLNOS AITL NKTCL EATL
Abbreviations: AITL, angioimmunoblastic Tcell lymphoma; BM, bone marrow; EATL, entheropathyassociated Tcell lymphoma; ENS, extranodal sites;
EPI, EATL prognostic index; IPI, International Prognostic Index; ECOG, Eastern Cooperative Oncology Group; IPTCLP, International Peripheral Tcell
Lymphoma Project; LDH, lactate dehydrogenase; mPIT, modified Prognostic Index for Tcell lymphoma; NKTCL, natural killer/Tcell lymphoma; PINK,
Prognostic Index of natural killer lymphoma; PLT, platelet count, PTCLNOS, peripheral Tcell lymphoma not otherwise specified; SAlb, serum albumin;
TCS, Tcell score.
54
-
LUMINARI AND SKRYPETS
Some recently published studies provide additional insights
into the role of autoSCT in CR1. A realworld data analysis from
the Swedish Lymphoma Registry found prolonged OS and PFS for
transplanted patients with PTCLNOS, AITL, ALKnegative ALCL
and EATCL after adjustment for confounding factors.
34
However,
the selection of nonASCT patients used as the control group may
have been biased by early progressing patients after induction
therapy. Another study was a large multicenter analysis conducted
by the LYSA group. Among the 527 studied cases, a final cohort of
269 patients age less than 65 years with a CR or PR after in-
duction chemotherapy was identified: 78 cases of PTCLNOS, 123
cases of AITL, and 68 cases of ALKnegative ALCL. Overall, 81%
were in CR and 19% in PR; 50% of the final cohort was allocated
to autoSCT (134 patients). Neither the Cox multivariate model nor
the propensity score analysis found any survival advantage in favor
of autoSCT as a consolidation procedure for patients in response
after induction therapy. Subgroup analyses did not reveal any
further difference in terms of response status, disease stage, or
risk category.
35
Recently, Park et al.
36
published their first report
of the large prospective observational COMPLETE study, con-
ducted by 56 U.S. academic centers. This paper described the
outcomes of 119 patients who achieved CR after induction ther-
apy, including 54 PTCLNOS, 35 AITL, and 30 ALKnegative ALCL.
Thirtysix patients underwent autoSCT; patients with AITL had
significantly improved OS and PFS but patients with other PTCL
subtypes did not. Finally, an exploratory of the ECHELON2 trial
was conducted, for the 82 patients with a declared intention to
transplant out of the 177 patients randomized to BVCHP arm
(ALKpositive ALCL were excluded). SCT was in fact performed in
38 patients (27 ALKnegative ALCL and 11 nonALCL patients),
most of whom were from nonAsian centers, suggesting regional
practice differences. Despite the fact that the ECHELON2 study
was not designed to evaluate the role of upfront consolidation
with ASCT, numerical PFS estimates favored the use of con-
solidative SCT in patients with ALKnegative ALCL and with non
ALCL who achieved a CR at the end of induction after frontline
BVCHP.
23
Interpreting the results from these studies on the role of
autoSCT consolidation is complicated by the diverse eligibility
criteria adopted, the suboptimal rates of transplantation among PTCL
subtypes, and the differing rates of CR before autoSCT. The decision
to proceed to autoSCT in a subject with PTCL who responds to first
line chemotherapy is difficult and should always be discussed with
the individual patient. Researchers are strongly encouraged to run
welldesigned clinical trials that adopt the same uptodate criteria
for response definition (i.e., FDGPET). These trials, which would
necessarily require considerable international cooperation, would
hopefully provide data by PTCL subtype.
alloSCT could be identified as alternative option to autoSCT as
consolidation of CR1 patients. Schmitz et al.
37
recently published
data from the first randomized phase 3 trial of auto versus alloSCT as
part of firstline therapy in poorrisk PTCLs excluding ALKpositive
TABLE 2ESMO and NCCN clinical practice guidelines for autoalloSCT in PTCLs
PTCLs subtype
Primary diagnosed PTCLs Relapsed/refractory PTCLs
ESMO NCCN ESMO NCCN
PTCLNOS PR, CR, transplant
eligible—autoSCT
Clinical trials, or ASCT, or
observation if CR, or
if PR—see rel/ref settings
PR, CR, transplant
eligible—alloSCT (or ASCT)
PR, CR, transplant
eligible—alloSCT (or ASCT)
AITL PR, CR, transplant
eligible—autoSCT
Clinical trials or ASCT or
observation if CR, or if
PR—see rel/ref settings
PR, CR, transplant
eligible—alloSCT (or ASCT)
PR, CR, transplant
eligible—alloSCT (or ASCT)
ALKnegative
ALCL
PR, CR, transplant
eligible—autoSCT
Clinical trials, or ASCT or
observation if CR, or
if PR—see rel/ref settings
PR, CR, transplant
eligible—alloSCT (or ASCT)
PR, CR, transplant
eligible—alloSCT (or ASCT)
ALKpositive
ALCL
No further treatment,
Or autoSCT
if highrisk profile
Only chemotherapy ±ISRT PR, CR, transplant
eligible—alloSCT (or ASCT)
PR, CR, transplant
eligible—alloSCT (or ASCT)
EATL autoSCT Clinical trials, or ASCT, or
observation if CR, or if
PR—see rel/ref settings
PR, CR, transplant
eligible—alloSCT (or ASCT)
PR, CR, transplant
eligible—alloSCT (or ASCT)
HSTCL ASCT or allo if
donor available
CR or PR—preferred alloSCT PR, CR, transplant
eligible—alloSCT (or ASCT)
Preferred alloSCT if eligible
ENKTCL ASCT Stage IV if CR—allo or ASCT PR, CR, transplant eligible—
alloSCT (or ASCT)
AlloSCT (or ASCT)
Abbreviations: AITL angioimmunoblastic Tcell lymphoma; ALCL, anaplastic large cell lymphoma; alloSCT, allogeneic stem cell transplant; ASCT,
autologous stem cell transplant; CR, Complete remission; EATL, entheropathyassociated Tcell lymphoma; ENKTCL, entranodal Tcell lymphoma;
HSTCL, hepatosplenic Tcell lymphoma; ISRT, involved site radiotherapy; PR, partial remission; PTCL, peripheral Tcell lymphoma.
LUMINARI AND SKRYPETS
-
55
ALCLs. The trial enrolled 18–60yearold patients of all stages and
IPI and was planned to detect an improvement of eventfree survival
at 3 years from 35% achieved with autoSCT to 60% by alloSCT in the
intenttotreat population. After the enrollment of 104 patient
randomization and recruitment was prematurely stopped because a
planned interim analysis had shown that it was highly unlikely to
meet the primary endpoint. The transplantrelated mortality
observed contributed to this result. In conclusion, alloSCT cannot be
recommended as consolidation therapy for CR1 PTCL patients due to
the lack of evidence and because of it toxicity profile. The only
TABLE 3Available prospective and retrospective studies of ASCT in PTCLs
Study NPTCLs subtype Time of transplant Response (%) PFS (years) OS (years)
Reimer et al. (2004) 83 39% PTCLNOS Upfront CR: 47 36% (4) 48% (4)
Prospective 16% ALCL (ALKnegative) PR: 24
33% AITL
Corradini et al. (2006) 62 45% PTCLNOS Upfront CR: 56 EFS reported:
30% (12)
34% (2)
Prospective 30% ALCL (ALKpositive) PR: 16
16% AITL
Rodriguez et al. (2007)
Prospective
26 42% PTCLNOS Upfront CR: 65 53% (3) 86% (3)
31% ALCL (ALKpositive) PR: 8
27% AITL
Mercadal et al. (2008) 41 49% PTCLNOS Upfront CR: 49 30% (4) 39% (4)
Prospective 29% AITL PR: 10
5% HSTCL
5% T/NK
d'Amore et al. (2012) 160 39% PTCLNOS Upfront CR: 83 44% (5) 51% (5)
Prospective 19% ALCL (ALKnegative) PR: 31
19% AITL
13% EATL
4% panniculitislike
3% T/NK
Fossard et al. (2018) 134 34% PTCLNOS Upfront CR: 75 46.3% (5) 59.2% (5)
Retrospective 23% ALCL (ALKnegative) PR: 25
43% AITL
Roerden et al. (2019) 58 25.9% AITL Upfront (40 pts) CR: 75 Upfront ASCT Upfront ASCT
22.4% EATCL Relapse/refractory
(18 pts)
PR: 25 44% (5) 53% (5)
20.7% PTCLNOS ASCT in first
relapse
ASCT in first
relapse
Retrospective 19% ALCL (ALKnegative) 60.6% (5) 77.4% (5)
8.6% ALCL (ALKpositive)
3.4% T/NK
Park et al. (2019) 36 42% PTCLNOS Upfront CR: 63 44% (5) 51% (5)
Prospective 11% ALCL (ALKnegative) PR: 37
47% AITL
Abbreviations: AITL, angioimmunoblastic Tcell lymphoma; ALCL, anaplastic large cell lymphoma; ALK, anaplastic lymphoma kinase negative; ASCT,
auto stem cell transplant; CR complete response; EATCL, enteropathyassociated Tcell lymphoma; EATL, entheropathyassociated Tcell lymphoma;
EFS, eventfree survival; HSTCL, hepatosplenic Tcell lymphomas; OS overall survival; PFS, progressionfree survival; PR, partial response; PTCLs,
peripheral Tcell lymphomas; PTCLNOS, peripheral Tcell lymphomas not otherwise specified; T/NK, Tcell lymphoma/natural killer.
56
-
LUMINARI AND SKRYPETS
exception to this general statement might be represented by hep-
atosplenic Tcell lymphomas for whom a systematic review of 44
cases treated with alloSCT at first or second relapse demonstrated a
3year relapsefree survival of 42% and OS of 56%.
38
1.3
|
Relapsed/refractory disease
Approximately 70% of patients with PTCL are expected to develop
relapsed or refractory disease after firstline therapy.
39,40
A dismal
outcome can be expected for these patients, with median OS of a few
months, even for those who are able to proceed to salvage therapy.
39
Among the options available, the effectiveness of autoSCT in
relapsed disease is uncertain due to the frequent use of autoSCT in
CR1 in eligible patients, and because the salvage therapies available
for relapsed PTCLs have very limited activity, thereby further
reducing the feasibility of an autoSCT program when planned.
Available salvage regimens were previously developed as a firstline
strategies mostly for Bcell malignancies, from them wellknown ICE
(ifosfamide, carboplatin, etoposide), DHAP (highdose cytarabine,
cisplatin, dexamethasone), GDP (gemcitabine, cisplatin, dexametha-
sone), and ESHAP (etoposide, cytarabine, cisplatin, and methylpred-
nisolone). Despite the fact that these regimens were previously
studied for aggressive lymphomas, due to the rarity of PTCLs a small
number of patients were included without independent subset
analysis.
41–44
Even if limited by a low power due, a subset analysis of
the Canadian Cancer Trials Group LY.12 randomized phase 3 study
was not able to confirm DHAP superiority over GDP in PTCLs.
45
In
relapsed/refractory PTCLs, alloSCT is also a feasible option in almost
all subtypes after failing prior autoSCT.
17,25,46
However, nonrelapse
mortality varies from 8.2% to 40%.
47–49
These scatter data suggest
that it is necessary to carefully select possible candidates for alloSCT.
Several new agents have been tested in the relapsed refractory
setting, and, while some have already received formal approval for
clinical use, approval is not uniform across countries. These include
the antiCD30 antibodydrug conjugate BV, pralatrexate (approved
in the United States only), and four histone deacetylase inhibitors
(HDAC): romidepsin (United States only), belinostat (United States
only), vorinostat (United States only), and chidamide (China only).
Results achieved with these agents are very similar with CR rates of
10%–25% and with a median PFS of less than 1 year.
1.4
|
Novel agents
Pharmacology research is very active in PTCL, and therapeutic
development is mainly driven by advances in the understanding of
the biology of the disease (Table 4).
TABLE 4Activity of novel agents from clinical trials in relapsed refractory PTCLs
Drug PTCL subtype No. of patients Study phase ORR; CR Reference
Pralatrexate PTCL 111 2 29%; 11% O'Connor et al. (2011)
Romidepsin PTCL 131 2 25%; 15% Coiffier et al. (2012)
Brentuximab vedotin CD30 +PTCL 34 2 41%; 24% Horwitz et al. (2012); Pro et al. (2012)
ALCL 58 86%; 57%
Belinostat PTCL 129 2 25.8%; 10.8% O'Connor et al. (2015)
Bendamustine PTCL 60 2 50%; 28% Damal et al. (2013)
Mogamulizumab CCR4 +PTCL/CTCL 29 1 34%; 17% Ogura et al. (2014)
Lenalidomide PTCL 39 2 26%; 8% Tournishey et al. (2015)
Copanlisib NHL 17 2 21%; 14% Dreyling et al. (2017)
Cerdulatinib PTCL 18 2a 43% Horwitz et al. (2018)
Duvelisib PTCL +CTCL 16 1 50%; 19% Horwitz et al. (2018)
Alisertib PTCL 271 3 33%; 18% O'Connor et al. (2019)
Tipifarnib AITL, CXCL12 +TCL 43 Interim analysis 45% Witzig et al. (2019)
Pembrolizumab PTCL 18 33%; 27% Barta et al. (2019)
Panobinostat +bortezomib PTCL 25 1 43%; 22% Tan et al. (2015)
Gemcitabine +romidepin PTCL 20 1 30%; 15% Pellegrini et al. (2016)
Pralatrexate +romidepsin TCL 14 1 71%; 29% Amengual et al. (2018)
5Azacytidine +romidepsin PTCL 31 1 73%; 55% O'Connor et al. (2019)
Abbreviations: AITL, angioimmunoblastic Tcell lymphoma; ALCL, anaplastic largecell lymphoma; CCR4, chemokine receptor4; CR, complete response;
CTCL, cutaneous Tcell lymphoma; CXCL12, CXC motif chemokine 12; NHL, nonHodgkin lymphoma; ORR, overall response rate; PTCL, peripheral T
cell lymphoma; TCL, Tcell lymphoma.
LUMINARI AND SKRYPETS
-
57
The frequent alteration of the epigenetic machinery in PTCL,
mainly of the TFH phenotype, justifies strong rationale for the search
of novel HDAC inhibitors and to test the efficacy of combining more
than one epigenetic modifier. A recent phase 1 combination trial of 5
azacitidine and romidepsin reported very interesting OR and CR
rates of 73% and 55%, respectively.
50
Inhibition of spleen tyrosine kinase (SYK) signaling and of PI3K
pathways have also been investigated, with promising results from
phase I and II studies. Cerdulatinib is an oral SYK, JAK1, JAK3, and
Tyk2 inhibitor; in a phase IIa study on 41 PTCL patients, it was able
to produce an overall response rate of 34%, with 27% CR rates.
51
Among PI3K inhibitors, the oral duvelisib was used in a phase I study
with 16 PTCL and 19 CTCL patients. The overall response rate was
50% for the PTCL patients and the median PFS was 8.4 months.
45
The same agent has been evaluated in combination with romidepsin,
showing greater activity in AITL and PTCLNOS (overall
response rates 74% and 64%, respectively, CR rates 63% and 36%,
respectively).
52
A promising therapeutic strategy in PTCL is represented by
targeting of tumor microenvironment. Blocking the PD1 interaction
with its ligand is justified by the finding of an increased expression of
PDL1 in both malignant and stromal cells of several PTCL subtypes.
Indeed, some activity of antiPD1 agents in PTCL has been described
by phase I studies,
53
and more convincing results have been achieved
with NKTCL. The use of PD1 blockers, however, has also been
associated with cases of hyperprogression, thus making further
clarification of PD1 inhibition in PTCL urgently needed.
Finally, cellular therapy based on the concept of chimeric antigen
receptor T cells is also being developed for Tcell lymphomas,
54
as is
the use of bispecific antibodies targeting both CD30 and CD16A.
55
2
|
CONCLUSIONS
Management of PTCL patients continues to be a real challenge for
hematologist–oncologists. The oversimplified approach that has been
used for many years, replicating the rules of Bcell lymphoma man-
agement, is clearly unsatisfactory and requires radical reassessment.
New insights into the biology of the disease and a renewed interest
on the part of the scientific community in the management of PTCL
have led to the identification of new targets and to confirming the
activity of new agents, which are now moving PTCLs into the era of
targeted therapy. Moreover, taking into account the different biology
and unique behavior of PTCL subtypes, each with a different
response to therapy, has become indispensable; these differences
result in more difficulties in interpreting the available data and in
designing future trials. PTCL remains a challenging disease which
requires massive international cooperation.
ACKNOWLEDGMENTS
The authors are grateful to MS Jacqueline Costa for her support in
reviewing the manuscript
CONFLICT OF INTERESTS
Luminari served as advisor for Roche, BMS, Jannsen, Regeneron,
Genmab, Gilead. Skrypets has no conflict of interest to declare.
ORCID
Stefano Luminari https://orcid.org/0000-0001-8446-2285
Tetiana Skrypets https://orcid.org/0000-0002-2276-869X
REFERENCES
1. Vose J, Armitage J, Weisenburger D, International TCell Lymphoma
Project. International peripheral Tcell and natural killer/Tcell
lymphoma study: pathology findings and clinical outcomes. J Clin
Oncol. 2008;26(25):41244130.
2. Bellei M, Chiattone CS, Luminari S, et al. Tcell lymphomas in South
America and Europe. Rev Bras Hematol Hemoter. 2011;34(1):4247.
https://doi.org/10.5581/15168484.20120013
3. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the
World Health Organization classification of lymphoid neoplasms.
Blood. 2016;127(20):23752390.
4. Bellei M, Foss FM, Shustov AR, et al. The outcome of peripheral
Tcell lymphoma patients failing first line therapy: a report from the
prospective, International TCell Project. Haematologica. 2018;
103(7):11911197.
5. Swerdlow SH, Campo E, Harris NL, et al. WHO Classification of Tu-
mours of Haematopoietic and Lymphoid Tissues. Lyon, France: Inter-
national Agency for Research on Cancer; 2017.
6. Heavican TB, Bouska A, Yu J, et al. Genetic drivers of oncogenic
pathways in molecular subgroups of peripheral Tcell lymphoma.
Blood. 2019;133(15):16641676. https://doi.org/10.1182/blood
201809872549
7. Wang T, Feldman AL, Wada DA, et al. GATA3 expression identifies
a highrisk subset of PTCL, NOS with distinct molecular and clinical
features. Blood. 2014;123(19):30073015. https://doi.org/10.1182/
blood201312544809
8. PinterBrown LC. Strategies for aggressive Tcell lymphoma: divide
and conquer. Hematology Am Soc Hematol Educ Program.
2020;2020(1):154159. https://doi.org/10.1182/hematology.20200
00101
9. Weisenburger DD, Savage KJ, Harris NL, et al. Peripheral Tcell
lymphoma, not otherwise specified: a report of 340 cases from the
International Peripheral Tcell Lymphoma Project. Blood. 2011;117:
34023408.
10. Went P, Agostinelli C, Gallamini A, et al. Marker expression in pe-
ripheral Tcell lymphoma: a proposed clinicalpathologic prognostic
score. J Clin Oncol. 2006;24:24722479.
11. Gallamini A, Stelitano C, Calvi R, et al. Peripheral Tcell lymphoma
unspecified (PTCLU): a new prognostic model from a retrospective
multicentric clinical study. Blood. 2004;103:24742479.
12. Schmitz N, Trümper L, Ziepert M, et al. Treatment and prognosis of
mature Tcell and NKcell lymphoma: an analysis of patients with
Tcell lymphoma treated in studies of the German HighGrade Non
Hodgkin Lymphoma Study Group. Blood. 2010;116:34183425.
13. Federico M, Bellei M, Marcheselli L, et al. Peripheral T cell lym-
phoma, not otherwise specified (PTCLNOS). A new prognostic
model developed by the International T Cell Project Network. Br J
Haematol. 2018;181(6):760769. https://doi.org/10.1111/bjh.15258
14. Cheson BD. PET/CT in lymphoma: current overview and future di-
rections. Semin Nucl Med. 2018;48(1):7681. https://doi.org/10.
1053/j.semnuclmed.2017.09.007
15. Pellegrini C, Argnani L, Broccoli A, et al. Prognostic value of interim
positron emission tomography in patients with peripheral Tcell
lymphoma. Oncologist. 2014;19:746750.
58
-
LUMINARI AND SKRYPETS
16. ElGalaly TC, Pedersen MB, Hutchings M, et al. Utility of interim and
endoftreatment PET/CT in peripheral Tcell lymphomas: a review
of 124 patients. Am J Hematol. 2015;90:975980.
17. d'Amore F, Gaulard P, Trumper L, et al. Peripheral Tcell lymphomas:
ESMO Clinical Practice Guidelines for diagnosis, treatment and
followup. Ann Oncol. 2015;26(suppl 5):v108v115.
18. AbouYabis AN, Shenoy PJ, Sinha R, Flowers CR, Lechowicz MJ. A
systematic review and metaanalysis of frontline anthracycline
based chemotherapy regimens for peripheral Tcell lymphoma.
ISRN Hematol. 2011;2011:623924. https://doi.org/10.5402/2011/
623924
19. Lage LAPC, Cabral TCS, Costa RO, et al. Primary nodal peripheral
Tcell lymphomas: diagnosis and therapeutic considerations. Rev
Bras Hematol Hemoter. 2015;37:277284.
20. Dearden CE, Johnson R, Pettengell R, et al. Guidelines for the
management of mature Tcell and NKcell neoplasms (excluding
cutaneous Tcell lymphoma). Br J Haematol. 2011;153:451485.
21. Karakas T, Bergmann L, Stutte HJ, et al. Peripheral Tcell lymphomas
respond well to vincristine, adriamycin, cyclophosphamide, predni-
sone and etoposide (VACPE) and have a similar outcome as high
grade Bcell lymphomas. Leuk Lymphoma. 1996;24(12):121129.
22. Wulf GG, Altmann B, Altmann B, et al. Alemtuzumab plus CHOP
versus CHOP in elderly patients with peripheral Tcell lymphoma:
the DSHNHL20061B/ACT2 trial. Leukemia. 2021;35(1):143155.
https://doi.org/10.1038/s4137502008385
23. Horwitz S, O'Connor OA, Pro B, et al. Brentuximab vedotin with
chemotherapy for CD30positive peripheral Tcell lymphoma
(ECHELON2): a global, doubleblind, randomised, phase 3 trial.
Lancet. 2019;393(10168):229240. https://doi.org/10.1016/S0140
6736(18)329842
24. Bachy E, Camus V, Thieblemont C, et al. Final analysis of the Ro
CHOP Phase III (conducted by LYSA): romidepsin plus CHOP in
patients with peripheral Tcell lymphoma. 62nd American Society of
Hematology (ASH) Annual Meeting and Exposition; 2020.
25. https://www.nccn.org/professionals/physician_gls/pdf/tcell_blocks.
pdf. Accessed 17 Feb 2021
26. Corradini P, Tarella C, Zallio F, et al. Longterm followup of patients
with peripheral Tcell lymphomas treated upfront with highdose
chemotherapy followed by autologous stem cell transplantation.
Leukemia. 2006;20:15331538.
27. Kyriakou C, Canals C, Goldstone A, et al. Highdose therapy and
autologous stemcell transplantation in angioimmunoblastic lym-
phoma: complete remission at transplantation is the major deter-
minant of outcome lymphoma working party of the European Group
for Blood and Marrow Transplantation. J Clin Oncol. 2008;26(2):
218224.
28. Rodriguez J, Conde E, Gutierrez A, et al. The adjusted International
Prognostic Index and β‐2microglobulin predict the outcome after
autologous stem cell transplantation in relapsing/refractory pe-
ripheral Tcell lymphoma. Haematologica. 2007;92:10671074.
29. Stiff PJ, Unger JM, Cook JR, et al. Autologous transplantation as
consolidation for aggressive nonHodgkin's lymphoma. N Engl J Med.
2013;369:16811690.
30. Mounier N, Gisselbrecht C, Brière J, et al. Prognostic factors in pa-
tients with aggressive nonHodgkin's lymphoma treated by front
line autotransplantation after complete remission: a cohort study
by the Groupe d'Etude des Lymphomes de l'Adulte. J Clin Oncol.
2004;22:28262834.
31. Mounier N, Gisselbrecht C, Brière J, et al. All aggressive lymphoma
subtypes do not share similar outcome after frontline auto-
transplantation: a matchedcontrol analysis by the Groupe d'Etude
des Lymphomes de l'Adulte (GELA). Ann Oncol. 2004;15:17901797.
32. d'Amore F, Relander T, Lauritzsen GF, et al. Upfront autologous
stemcell transplantation in peripheral Tcell lymphoma: NLGT01.
J Clin Oncol. 2012;30:30933099.
33. Reimer P, Schertlin T, Rüdiger T, et al. Myeloablative radio-
chemotherapy followed by autologous peripheral blood stem cell
transplantation as firstline therapy in peripheral Tcell lymphomas:
first results of a prospective multicenter study. Hematol J.
2004;5(4):304311. https://doi.org/10.1038/sj.thj.6200359
34. Ellin F, Landström J, Jerkeman M, Relander T. Realworld data on
prognostic factors and treatment in peripheral Tcell lymphomas: a
study from the Swedish Lymphoma Registry. Blood. 2014;
124(10):15701577.
35. Fossard G, Broussais F, Coelho I, et al. Role of upfront autologous
stemcell transplantation in peripheral Tcell lymphoma for patients
in response after induction: an analysis of patients from LYSA cen-
ters. Ann Oncol. 2018;29(3):715723. https://doi.org/10.1093/
annonc/mdx787
36. Park SI, Horwitz SM, Foss FM, et al. The role of autologous stem cell
transplantation in patients with nodal peripheral Tcell lymphomas
in first complete remission: report from COMPLETE, a prospective,
multicenter cohort study. Cancer. 2019;125(9):15071517. https://
doi.org/10.1002/cncr.31861
37. Schmitz N, Truemper LH, Bouabdallah K, et al. A randomized phase 3
trial of auto vs. allo transplantation as part of firstline therapy in
poorrisk peripheral TNHL. Blood. 2020. https://doi.org/10.1182/
blood.2020008825
38. Rashidi A, Cashen AF. Outcomes of allogeneic stem cell trans-
plantation in hepatosplenic Tcell lymphoma. Blood Cancer J.
2015;5(6):e318. https://doi.org/10.1038/bcj.2015.43
39. Mak V, Hamm J, Chhanabhai M, et al. Survival of patients with
peripheral Tcell lymphoma after first relapse or progression:
spectrum of disease and rare longterm survivors. J Clin
Oncol. 2013;31(16):19701976. https://doi.org/10.1200/JCO.2012.
44.7524
40. Dreyling M, Thieblemont C, Gallamini A, et al. ESMO Consensus
conferences: guidelines on malignant lymphoma. Part 2: marginal
zone lymphoma, mantle cell lymphoma, peripheral Tcell lymphoma.
Ann Oncol. 2013;24:857877.
41. Jerkeman M, Leppä S, Kvaløy S, Holte H. ICE (ifosfamide, carbo-
platin, etoposide) as secondline chemotherapy in relapsed or pri-
mary progressive aggressive lymphoma—the Nordic Lymphoma
Group experience. Eur J Haematol. 2004;73:179182.
42. Philip T, Chauvin F, Armitage J, et al. Parma international protocol:
pilot study of DHAP followed by involvedfield radiotherapy and
BEAC with autologous bone marrow transplantation. Blood.
1991;77:15871592.
43. Zelenetz AD, Hamlin P, Kewalramani T, Yahalom J, Nimer S,
Moskowitz CH Ifosfamide, carboplatin, etoposide (ICE)based
secondline chemotherapy for the management of relapsed and re-
fractory aggressive nonHodgkin's lymphoma. Ann Oncol.
2003;14(suppl 1):i5i10.
44. Velasquez WS, McLaughlin P, Tucker S, et al. ESHAP—an effective
chemotherapy regimen in refractory and relapsing lymphoma: a 4
year followup study. J Clin Oncol. 1994;12:11691176.
45. Skamene T, Crump M, Savage KJ, et al. Salvage chemotherapy and
autologous stem cell transplantation for peripheral Tcell lymphoma:
a subset analysis of the Canadian Cancer Trials Group LY.12 ran-
domized phase 3 study. Leuk Lymphoma. 2017;58(10):23192327.
https://doi.org/10.1080/10428194.2017.1312379
46. KharfanDabaja MA, Kumar A, Ayala E, et al. Clinical practice rec-
ommendations on indication and timing of hematopoietic cell
transplantation in mature T cell and NK/T cell lymphomas: an in-
ternational collaborative effort on behalf of the guidelines commit-
tee of the American Society for Blood and Marrow Transplantation.
Biol Blood Marrow Transplant. 2017;23(11):18261838. https://doi.
org/10.1016/j.bbmt.2017.07.027
47. Loirat M, Chevallier, P, Leux C, et al. Upfront allogeneic stemcell
transplantation for patients with nonlocalized untreated peripheral
LUMINARI AND SKRYPETS
-
59
Tcell lymphoma: an intentiontotreat analysis from a single center.
Ann Oncol. 2015;26(2):386392.
48. Kyriakou C, Canals C, Finke J, et al. Allogeneic stem cell
transplantation is able to induce longterm remissions in
angioimmunoblastic Tcell lymphoma: a retrospective study from
the lymphoma working party of the European group for blood
and marrow transplantation. J Clin Oncol. 2009;27(24):
39513958.
49. Smith SM, Burns LJ, van Besien K, et al. Hematopoietic cell trans-
plantation for systemic mature Tcell nonHodgkin lymphoma. J Clin
Oncol. 2013;31(25):31003109.
50. O'Connor OA, Falchi L, Lue JK, et al. Oral 5azacytidine and romi-
depsin exhibit marked activity in patients with PTCL: a multicenter
phase 1 study. Blood. 2019;134(17):13951405. https://doi.org/10.
1182/blood.2019001285
51. Horwitz SM, Feldman TA, Hess BT, et al. The novel SYK/JAK in-
hibitor cerdulatinib demonstrates good tolerability and clinical
response in a phase 2a study in relapsed/refractory peripheral Tcell
lymphoma and cutaneous Tcell lymphoma. Blood. 2018;132(sup-
plement 1):1001. https://doi.org/10.1182/blood201899119944
52. Horwitz SM, Koch R, Porcu P, et al. Activity of the PI3K‐δ,γinhibitor
duvelisib in a phase 1 trial and preclinical models of Tcell
lymphoma. Blood. 2018;131(8):888898. https://doi.org/10.1182/
blood201708802470
53. Lesokhin AM, Ansell SM, Armand P, et al. Nivolumab in patients with
relapsed or refractory hematologic malignancy: preliminary results
of a phase Ib study. J Clin Oncol. 2016;34(23):26982704. https://doi.
org/10.1200/JCO.2015.65.9789
54. Maciocia PM, Wawrzyniecka PA, Philip B, et al. Targeting the T cell
receptor β‐chain constant region for immunotherapy of T cell ma-
lignancies. Nat Med. 2017;23(12):14161423. https://doi.org/10.
1038/nm.4444
55. Sawas A, Elgedawe H, Vlad G, et al. Clinical and biological evaluation
of the novel CD30/CD16A tetravalent bispecific antibody (AFM13)
in relapsed or refractory CD30positive lymphoma with cutaneous
presentation: a biomarker phase Ib/IIa study (NCT03192202). Blood.
2018;132(supplement 1):2908.
How to cite this article: Luminari S, Skrypets T. What's new in
peripheral Tcell lymphomas. Hematological Oncology.
2021;39(S1):5260. https://doi.org/10.1002/hon.2846
60
-
LUMINARI AND SKRYPETS
Received: 26 February 2021
DOI: 10.1002/hon.2854
SUPPLEMENT ARTICLE
Recognizing but not harming. Borderline Bcell lymphoid
proliferations
Leticia QuintanillaMartinez
Institute of Pathology and Neuropathology,
Eberhard Karls University of Tübingen and
Comprehensive Cancer Center, Tübingen
University Hospital, Tübingen, Germany
Correspondence
Leticia QuintanillaMartinez, Institute of
Pathology, University Hospital Tübingen
Liebermeisterstrasse 8, 72076 Tübingen,
Germany.
Email: Leticia.quintanilla-fend@med.uni-
tuebingen.de
Funding information
Open access funding was enabled and
organized by Projekt DEAL.
Abstract
In the last years several borderline Bcell lymphoid proliferations have been recog-
nized that lie at the interface between benign and malignant. These lesions can be
divided in two groups; those considered precursor lesions of well recognized lymphoid
malignancies and the group of indolent lymphomas with limited potential for pro-
gression. Precursor lesions are monoclonal and share many genetic features of their
malignant counterpart. The first recognized precursor lesion was monoclonal
gammopathy of undetermined significance. Thereafter, the widespread use of
immunohistochemistry, fluorescenceactivated cell sorting analysis and molecular
techniques in lymphoid samples led to the recognition of other precursor lesions such
as monoclonal Bcell lymphocytosis, in situ follicular neoplasia and in situ mantle cell
neoplasia. The second group of disorders comprises monoclonal lymphoid pro-
liferations with limited malignant potential without a counterpart among the
currently recognized lymphoma entities; these include pediatrictype follicular lym-
phoma, pediatric nodal marginal zone lymphoma, and duodenaltype follicular lym-
phoma. Despite their clonal nature, a conservative treatment has been shown to be
sufficient in most cases. The diagnostic criteria of precursor/indolent Bcell pro-
liferations, recent advances in the understanding of progression and lymphoma-
genesis and current recommendations for treatment will be discussed. In order to
avoid unnecessary and potentially harmful therapy, these lesions need to be recog-
nized and diagnosed correctly.
KEYWORDS
Bcell lymphoma, early lesion, indolent lymphoma, precursor lesion
1
|
INTRODUCTION
The paradigm that invasive cancer is frequently preceded by pre-
malignant lesions has been proven in solid tumors. However, this
concept is difficult to apply to lymphoid neoplasms because lym-
phocytes due to their innate properties recirculate through different
lymphoid tissues and organs based on the patterns of normal
lymphocyte homing.
1
The increasing use of immunohistochemistry, fluorescence
activated cell sorting (FACS) analysis and molecular techniques on
lymphoid tissue samples has led to the identification of a number of
borderline Bcell lymphoid proliferations between benign and
This is an open access article under the terms of the Creative Commons AttributionNonCommercialNoDerivs License, which permits use and distribution in any
medium, provided the original work is properly cited, the use is noncommercial and no modifications or adaptations are made.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
Hematological Oncology. 2021;39(S1):6167. wileyonlinelibrary.com/journal/hon
-
61
malignant.
1
Some of these lesions represent precursor lesions and
some clonal lymphoid proliferations with limited malignant potential
(Table 1). The precursor lymphoid proliferations are often incidental
findings in asymptomatic individuals with mostly indolent biological
behavior. Importantly, these lymphoid proliferations are monoclonal
and carry many of the molecular features of their malignant coun-
terpart such as the t(14;18) and t(11;14) translocations associated to
in situ follicular neoplasia (ISFN) and in situ mantle cell neoplasia
(ISMCN), respectively. Other Bcell proliferations with low risk of
progression include monoclonal Bcell lymphocytosis (MBL) of
chronic lymphocytic leukemia (CLL)type and the nonCLL type. In
pediatric and young populations two clonal Bcell proliferations with
apparently no risk of progression are recognized including pediatric
type follicular lymphoma (PTFL) and pediatric nodal marginal zone
lymphoma (PMZL). There are some lymphoid proliferations that
home to specific anatomic sites and remain localized such as
duodenaltype follicular lymphoma (DFL).
2
|
PRECURSOR LESIONS IN BCELL
LYMPHOMAGENESIS
2.1
|
Monoclonal gammopathy of undetermined
significance
Monoclonal gammopathy of undetermined significance (MGUS) is the
paradigmatic example for the clonal evolution of lymphoid neo-
plasms.
2
It is defined as a serum monoclonal protein less than 3 g/dl,
less than 10% monoclonal plasma cells in the bone marrow (BM) and
absence of SLiMCRAB criteria (60% clonal plasma cells, free light
chain (FLC) ratio 100, >1 focal lesion on magnetic resonance im-
aging, hypercalcemia, renal insufficiency, anemia and lytic bone le-
sions). MGUS is a premalignant clonal disorder that progresses at a
rate of 1% per year. The incidence of MGUS increases with age
affecting 3%–4% individuals 50 years and 5% at 70 years. It is
classified based on the involved immunoglobulin (IG) in IGM, non
IGM and light chain types. NonIGM MGUS might progress to mul-
tiple myeloma (MM), solitary plasmacytoma or amyloid light chain
disease, whereas IGM MGUS has an increased risk to progress to
Waldeström macroglobulinemia (WM) or CLL. Interestingly, MGUS,
smoldering MM (SMM) and MM share cytogenetic alterations
including major translocations involving the IGH gene and numerical
chromosomal abnormalities (CNAs). Nevertheless, there seems to be
a temporal acquisition of CNAs, some of which are more prevalent at
later stages such as del13q, indicating that they may be secondary
events. Similarly, although t(11;14) is observed in MGUS and MM, t
(4;14) and t(14;16) are more common in SMM and MM. MGUS has a
less complex genomic landscape carrying rarely MMassociated so-
matic mutations. Furthermore, MYC translocations and TP53 de-
letions and mutations are not observed in MGUS, suggesting that
these are drivers of progression. The most common CNAs for IGM
MGUS/WM are del6q, +18q, trisomy 4, 5, 12, and monosomy 8. The
most frequent mutation is MYD88 L265P, followed by mutations in
CXCR4 and KMT2D, akin to manifest WM.
A major advance in the understanding of MM pathogenesis was
the recognition of the dynamic interaction between the tumor cells
TABLE 1Precursor/indolent lesions
in lymphoid neoplasias
Precursor lesions phenotypically and genotypically similar to malignant counterpart
Monoclonal gammopathy of undetermined significance
Monoclonal Bcell lymphocytosis
Chronic lymphocytic leukemia (CLL) type
Atypical CLL type
NonCLL type
In situ follicular neoplasia
In situ mantle cell neoplasia
Clonal lymphoid proliferations of limited malignant potential (B and T)
Duodenaltype follicular lymphoma
Pediatrictype follicular lymphoma
Pediatric nodal marginal zone lymphoma
Primary cutaneous marginal zone lymphoma
Primary cutaneous acral CD8+Tcell lymphoma
Primary cutaneous CD4+Tcell lymphoproliferative disorder (LPD)
Indolent Tcell LPD of the gastrointestinal tract
Lymphomatoid papulosis
62
-
QUINTANILLAMARTINEZ
and the surrounding microenvironment. The permissive BM micro-
environment favors clonal selection and progression from MGUS to
MM. The BM microenvironment seems to exert a selective pressure
that promotes the expansion of subclones with selective advantage
and facilitates progression to symptomatic disease. Additionally, the
immunosuppressive BM microenvironment through expansion of
regulatory Tcells and Thelper 17 cells determines whether MGUS
remains stable or progresses to MM.
There are several risk stratification models to predict MGUS
progression. The most widely used is the “Mayo Clinic model”: (1) M
protein greater than 1.5 g/dl, (2) nonIgG isotype, and (3) FLC ratio
less than 0.26 or more than 1.65, which stratifies patients from low to
high risk. High risk patients have all three risk factors and a 20years
progression risk of 58%. Currently, there is no recommended treat-
ment for MGUS except for careful monitoring for progression to
enable early detection and intervention.
3
An exception is what has
been called “MGCS.” This refers to patients with MGUS criteria but
experiencing symptoms and signs related to the nerves, skin and
kidney causing serious disease. Treatment recommendations include
IV IG, rituximab, and plasma celldirected therapy.
4
There are
promising clinical trials investigating different therapeutic strategies
to prevent progression from highrisk MGUS to symptomatic disease
including a phase II trial with daratumumab.
3
Introduction of novel
therapies that seem to be both safe and effective are challenging the
belief that MM should only be treated when symptomatic.
2.2
|
Monoclonal Bcell lymphocytosis
MBL is defined as a monoclonal Bcell count less than 5000/μl in
peripheral blood (PB), persisting for at least 3 months in otherwise
asymptomatic individuals.
5
Based on its clinical significance, MBL is
divided into highcount (500–5000/µl) and lowcount MBL (<500/
µl); the latter without risk to develop CLL, and therefore, no
followup is required. MBL is classified into three categories based
on its phenotype: (1) CLL type, (2) atypicalCLL type, and (3) non
CLL type.
(1) MBL of CLL type is the most common (75%), and has an identical
phenotype to CLL (CD19+, CD20dim, CD23+, CD5+, LEF1+).
Among healthy individuals, MBL has a reported incidence be-
tween 3% and 12%, depending on the age. CLL is always pre-
ceded by MBL, which is considered a premalignant condition
with a progression rate to CLL of 1%–2% per year; however, only
a minority of MBL cases will eventually progress into CLL.
1
The
number of MBL cells at presentation seems to be the highest risk
factor for progression. It is now well established that progression
from MBL to CLL is due to the combined effect of genetic defects
and its interaction with the microenvironment.
1
MBL cases
usually have mutated IGHV, and carry the same genetic alter-
ations as seen in CLL including del13q, trisomy 12, del11q, and
del17p, although at lower levels suggesting that these changes
occur early in clonal evolution. Point mutations in TP53,
NOTCH1,ATM, and SF3B1 genes are also detected in MBL but
mostly as subclones.
6,7
Current recommendations include yearly
monitoring for early intervention if clinically indicated.
(2) MBL with atypical CLL phenotype shows CD20 bright by FACS
analysis and often is CD23 negative but express CD5. The pos-
sibility of a mantle cell lymphoma (MCL) should be excluded.
(3) MBL of nonCLL type is characterized by CD20+, CD19+B cells
with moderate to bright surface IG expression but CD5 nega-
tivity (or CD5 dim in 20% of cases), a phenotype that has been
related to marginal zone (MZ) lymphoma. In PB the lymphocytes
show a broad cytological spectrum with variable numbers of
villous lymphocytes and lymphocytes with plasmacytoid differ-
entiation. BM biopsy shows always Bcell infiltrations of various
degrees. Most patients remain stable without progression but
around 17% will develop splenomegaly and laboratory findings
suggestive of splenic MZ lymphoma or splenic Bcell lymphoma
unclassifiable. The term CBLMZ origin has been proposed.
8
The
immunogenetic signature of these cases is strongly indicative of
antigen selection with highly mutated IG genes (76%) and pre-
dominance of the IGHV434 gene (23%). Up to 23% of the cases
show complex karyotype with frequent del7q and isochromo-
some 17q. Mutational analysis is limited to few cases; however,
NOTCH2 (13%), TNFAIP3 (6%), and CD79B (6%) gene mutations
have been described.
9
Interestingly, MYD88 mutations were
identified in 25% of cases, especially those cases secreting IGM;
the differential diagnosis with WM is mandatory. Accordingly, it
is recommended to evaluate individuals at diagnosis with
detailed immunophenotype, cytogenetics (fluorescence in situ
hybridization), MYD88 gene mutation screening and imaging. The
role of BM biopsy at diagnosis is debatable since there is no
correlation between the extent of BM and PB infiltration. Long
followup is recommended but treatment should only be
considered in cases with evidence of progression.
2.3
|
In situ follicular neoplasia
FL is the second most common Bcell lymphoma in adults and ac-
counts for about 20% of all lymphomas. Most patients have wide-
spread disease at diagnosis with BM involvement in 40%–70% of the
cases. FL cells acquire resistance against apoptosis through the
chromosomal translocation t(14;18)/BCL2/IGH present in around
85% of the cases. Using immunohistochemistry and highly sensitive
molecular techniques two putative FL precursor forms have been
identified.
10
(1) FLlike B cells carrying the t(14;18) translocation can
be identified in the PB in up to 55% of healthy individuals. These
cells increase with age, smoking, and exposure to pesticides. Most
t(14;18)positive clones are longlived and can persist for several
years, indicating that the deregulated BCL2 expression is not suffi-
cient for the development of manifest (m)FL. FLlike B cells have high
levels of somatic hypermutation (SHM) and class switch recombina-
tion (CSR) suggestive of intense trafficking in the germinal center
(GC). Importantly, patients who develop mFL seem to have a higher
QUINTANILLAMARTINEZ
-
63
number of circulating FLlike Bcells many years before the diagnosis
of mFL compared with controls.
(2) ISFN is a precursor lesion considered to be a tissue manifes-
tation of FLlike B cells, which show the genetic and immunopheno-
typic features of FL, but localized to the GC of reactiveappearing
lymph nodes (LN) (Figure 1). ISFN can only be identified by BCL2
immunostain, is monoclonal and carries the t(14;18) translocation. The
incidence of ISFN is unknown, but studies of unselected LNs from
adults revealed a prevalence of 2%, and the rate for progression to mFL
has been estimated to be between 2% and 3%. Nevertheless, because
15%–20% of ISFN cases are diagnosed in a LN involved by another B
cell lymphoma (not only FL), it is recommended to have complete
staging workup, including imaging studies and BM biopsy to rule out
concurrent lymphoma. No treatment is needed.
11
ISFN represents a FL precursor lesion with no or few CNAs.
11
Nevertheless ISFN shares many genetic features with mFL including
early introduction of Nglycosylation sites substituting conventional
antigen binding favoring the generation of longlived clones and
increasing the chances of secondary hits. Accordingly, ISFN carry
mutations in CREBBP,EZH2,TNFRSF14, and KMT2D but at lower
frequencies when compared to paired mFL.
12
The putative model of FL
lymphomagenesis suggests that the t(14;18) is the first hit resulting
from a repair error during the IGH variable diversity joining region
recombination process in the BM leading to the accumulation of cells
that will undergo repetitive rounds of GC reaction, acquiring additional
hits by SHM and CSR (FLlike B cells). These cells acquire functional N
glycosylation sites that will interact with M2polarized macrophages in
the GC that retain the cells in the GC (ISFN), demonstrating the
importance of the interaction of activated Bcell receptor (BCR) and its
microenvironment. These cells will undergo clonal expansion and
evolution with acquisition of CNAs and chromatin modifier gene mu-
tations necessary for final transformation into mFL. In contrast to
MGUS and MBL, there are no predictive biomarkers available to know
which individuals will progress from ISFN to mFL.
2.4
|
In situ mantle cell neoplasia
MCL represents around 6%–9% of all Bcell lymphomas. At diagnosis
most patients present with disseminated disease with frequent
FIGURE 1 Precursor lesions in Bcell lymphomas. In situ mantle cell neoplasia. (A) Panoramic view of a normalappearing lymph node
(hematoxylin and eosin [H&E], 25). (B) Cyclin D1 highlights the presence of cyclin D1+cells in the inner cuff of the mantle zone
(immunohistochemistry, 25). (C) The cells are also positive for the transcription factor SOX11 (immunohistochemistry, 200). (D) Higher
magnification demonstrates a specific nuclear staining for cyclin D1 (immunohistochemistry, 200). Insert: Fluorescence in situ hybridization
(FISH) analysis using a CCND1 breakapart probe (BAP) demonstrates one normal colocalized signal (yellow arrow) and two separate signals
(red and green arrows) indicative of a translocation. In situ follicular neoplasia. (E) Panoramic view of a lymph node stain with H&E showing
normal architecture (original magnification, 25). (F) BCL2 stain shows that several germinal centers are replaced by lymphoid cells with
strong BCL2 expression (immunohistochemistry, 25). Insert: FISH analysis using a BCL2 BAP demonstrates one normal colocalized signal
(yellow arrow) and two separate signals (red and green arrows) indicative of a translocation. (G) The cells are strongly positive for CD10.
(H) MIB1 stain shows an abnormally low proliferation in the upper follicle with a normal polarized proliferation in the lower follicle. (I) BCL2
stain shows the abnormal expression of BCL2 in the upper follicle, stronger than the normal B cells of the mantle zone and reactive T cells. In
contrast in the lower follicle the germinal center remains BCL2 negative (G–I, immunohistochemistry, 200)
64
-
QUINTANILLAMARTINEZ
infiltration of the gastrointestinal tract and BM. MCL is characterized
by the t(11;14)/CCND1/IGH translocation in more than 90% of cases
with the residual 10% showing mostly cryptic CCND2/IG or CCND3/
IG translocations.
6
The phenotype includes the expression of Bcell
markers (CD19, CD20) together with expression of CD5 and
SOX11 in most nodal cases. Rare cells with t(11;14) translocation
have been detected in up to 7% of healthy individuals in PB con-
firming that the CCND1IGH translocation is an early event in mantle
cell lymphomagenesis. MCLlike cells and ISMCN are seen at a
significantly lower frequency than precursor lesions in FL. ISMCN is a
rare incidental finding in lymphoid tissue (<1%) characterized by the
presence of atypical cyclin D1+cells located within the 3–5 inner
layers of the mantle cuffs in an otherwise normal LN (Figure 1).
ISMCN is less characterized than ISFN; however, these two precur-
sor lesions seem to have in common the dependency on activated
BCR and the interaction with the microenvironment crucial for
homing to their natural niche.
1
The prolonged and repeated antigen
stimulation is believed to trigger the accumulation of genetic and
epigenetic aberrations in the MCL precursor cells, which eventually
progress to an overt lymphoma. Current recommendations include
staging with imaging and BM biopsy to rule out a concurrent overt
lymphoma. No treatment is required and followup is advised. The
risk to progression is unknown.
3
|
INDOLENT LOCALIZED BCELL LYMPHOMAS
3.1
|
Duodenaltype follicular lymphoma
DFL is a FL variant with unique clinical features. Patients present
with small polypoid lesions confined to the mucosa and submucosa
often in the region of the ampulla of Vater but can also involve
jejunum, ileum, and even colon. The clinical course is indolent with a
risk of progression and dissemination of about 3%. The overall sur-
vival at 10 years is 100% with limited or no treatment. The diagnosis
of DFL is often incidental and detected upon endoscopy for other
unrelated causes.
13
Histologically, the lesion is characterized by well
circumscribed follicles composed almost entirely by centrocytes that
also infiltrate the lamina propria with the same phenotype as FL
(CD10+, BCL2+, and BCL6+) (Figure 2). Some studies have shown
disrupted follicular dendritic cell meshworks and reduced or lack
activationinduced deaminase expression compared to nodal FL. DFL
is monoclonal and carries the t(14;18) translocation. The expression
of α4β7 integrin by the tumor cells seem to be responsible for the
peculiar homing of these cells to the intestinal mucosa, a homing
receptor observed in mucosaassociated lymphoid tissue lymphomas.
Similar to ISFN, DFL has low genomic complexity with frequent
mutations in CREBBP,TNFRSF14,KMT2D, and EZH2. Multiple KMT2D
mutations as seen in advanced FL are absent. Gene expression
profiling suggested that the immune microenvironment of DFL is
distinct from nodal FL and characterized by a chronic inflammatory
signature relevant to the pathophysiology of this disease. DFL and
ISFN are considered early/precursor lesions, clinically indolent with
low risk of progression. In fact DFL most probably represents a form
of in situ FL of the gastrointestinal tract. The Lugano staging system
is recommended to evaluate the potential of progression. If nodal
involvement is present (Stage II) there is a higher rate of progression.
DFL should be distinguished from conventional FL with intestinal
involvement. In such cases, mesenteric LN involvement is often
present. There is no significant difference in survival or time to
progression between treated and untreated patients, which supports
the view that “watch and wait” strategy is preferred to therapeutic
intervention.
14
3.2
|
Pediatrictype follicular lymphoma
PTFL is a FL variant that differs clinically, morphologically, and
genetically from the adult counterpart. Although this lymphoma
predominates in children and adolescents, similar cases have been
described in adults.
15
PTFL is recognized as an indolent disease
affecting predominantly young males (M:F ratio of 20:1) with a me-
dian age of 15 years, and presenting with isolated lymphadenopathy
in the head and neck regions.
16
Morphologically, it is characterized by
partial or total LN effacement by expansile, serpiginous follicles with
a prominent “starrysky” pattern but without polarization. In some
cases, evidence of MZ differentiation may be seen in the periphery of
the neoplastic follicles. The cells within the follicles show highgrade
cytology (Grade 3) and express strongly CD10 and BCL6. BCL2
expression is usually absent, but around 20% of the cases show weak
positivity. PTFL is always monoclonal, which is required for the
diagnosis to exclude “atypical” follicular hyperplasia. The tumor cells
lack the characteristic t(14;18)/BCL2/IGH translocation of conven-
tional FL. Overall, PTFL lacks mutations of histone modifying genes
frequently found in FL and shows low levels of genomic complexity.
17
Recurrent genetic alterations of potential importance for its patho-
genesis that disrupt pathways associated with the GC reaction
(TNFRSF14,IRF8), immune escape (TNFRSF14) and antiapoptosis
(MAP2K1) have been described.
15,18
Complete staging involving BM
biopsy and lumbar puncture is standard care for children. Most pa-
tients presenting with localized disease have an excellent response
with surgery alone (completed resected lesion) without further
therapy. The actual recommendation is “watch and wait” after
resection. For patients with limited stage disease without complete
resection two to four courses chemotherapy according to protocols
for children with mature Bcell lymphomas is the standard of care.
Relapses are very rare. The overall survival of children and adoles-
cents with PTFL is 100% in all published series.
19
3.3
|
Pediatric nodal MZ lymphoma
PMZL is uncommon and shows remarkably overlapping clinical and
pathological features with PTFL. It is more common in males under
18 years of age and presents as asymptomatic isolated lymphade-
nopathy commonly in the head and neck region. Histologically,
QUINTANILLAMARTINEZ
-
65
residual follicles in PMZL are often present and show fragmentation
due to infiltration of the mantle cells into the remaining GC resem-
bling progressive transformation of germinal centers (PTGC). The
mantle zone is expanded and can be highlighted by IgD staining,
which in addition, emphasizes the PTGClike changes. In contrast to
PTFL, there is interfollicular infiltration of CD10, BCL6tumor cells
and diffuse areas. Staining for CD279/PD1 shows numerous cells in
the reactive GCs, a feature that is useful in the differential diagnosis
with PTFL. Monoclonality is detected in all cases. No recurrent mu-
tations have been identified. These cases show low levels of genomic
complexity with few genetic aberrations such as trisomy 18 and tri-
somy 3. Similar to PTFL a “watch and wait” strategy is recommended
after resection.
19
4
|
CONCLUSIONS
The precursor lesions discussed here have unique clinical, morpho-
logical, immunophenotypical, and genetic features. Nevertheless,
MBL, ISFN, and ISMCN share some biological features, such as
dependency on BCR activation and the important interaction with
the microenvironment. These precursor lesions can be detected in
healthy individuals and in general have a low risk of progression to a
malignant lymphoma. The complex interplay between genetic alter-
ations, BCR activation and microenvironment seems to be respon-
sible for progression to an overt lymphoma. Patients with these
precursor lesions should be staged at diagnosis but treatment is
recommended only if progression occurs, although this dogma might
change in the future. Indolent lymphomas such as DFL, PTFL, and
PMZL represent challenging diagnoses sometimes difficult to sepa-
rate from reactive conditions. The understanding and definition of
these disorders have increased tremendously in the last 5 years. For
PTFL and PMZL complete surgical resection followed by “watch and
wait” strategy seems an adequate therapy with lowintensity
chemotherapy restricted to incompletely resected disease. Finally,
the most important take home message is to recognize these lesions
and avoid unnecessary and potentially harmful therapy.
ACKNOWLEDGMENTS
Open access funding was enabled and organized by Projekt DEAL.
FIGURE 2 Duodenaltype follicular lymphoma. (A) Aggregates of neoplastic follicles mainly restricted to the duodenal mucosa
(hematoxylin and eosin [H&E], snapshot of a scanned slide). (B) Strong and homogeneous CD20 positivity in the follicular structures and in the
lamina propria (immunohistochemistry, snapshot of a scanned slide). (C) At higher magnification the follicles are replaced by mainly
centrocytelike cells (Grade 1) (H&E, 100). (D) CD23 stain highlights the abnormal distribution of follicular dendritic cells in the periphery of
the neoplastic follicles (immunohistochemistry, 100) The tumor cells are positive for (E) BCL2, (F) BCL6, (G) and CD10
(immunohistochemistry, 50)
66
-
QUINTANILLAMARTINEZ
CONFLICT OF INTERESTS
The author declares that there is no conflict of interests.
PEER REVIEW
The peer review history for this article is available at https://publons.
com/publon/10.1002/hon.2854.
REFERENCES
1. Ghia P, Nadel B, Sander B, Stamatopoulos K, Stevenson FK. Early
stages in the ontogeny of small Bcell lymphomas: genetics and
microenvironment. J Intern Med. 2017;282(5):395414.
2. Mouhieddine TH, Weeks LD, Ghobrial IM. Monoclonal gammopathy
of undetermined significance. Blood. 2019;133(23):24842494.
3. Ho M, Patel A, Goh CY, Moscvin M, Zhang L, Bianchi G. Changing
paradigms in diagnosis and treatment of monoclonal gammopathy of
undetermined significance (MGUS) and smoldering multiple
myeloma (SMM). Leukemia. 2020;34(12):31113125.
4. Dispenzieri A. Monoclonal gammopathies of clinical significance.
Hematology Am Soc Hematol Educ Program. 2020;2020(1):380388.
5. Marti GE, Rawstron AC, Ghia P, et al. Diagnostic criteria for mono-
clonal Bcell lymphocytosis. Br J Haematol. 2005;130(3):325332.
6. Puente XS, Jares P, Campo E. Chronic lymphocytic leukemia and
mantle cell lymphoma: crossroads of genetic and microenvironment
interactions. Blood. 2018;131(21):22832296.
7. Ojha J, Secreto C, Rabe K, et al. Monoclonal Bcell lymphocytosis is
characterized by mutations in CLL putative driver genes and clonal
heterogeneity many years before disease progression. Leukemia.
2014;28(12):23952398.
8. Xochelli A, Kalpadakis C, Gardiner A, et al. Clonal Bcell lymphocy-
tosis exhibiting immunophenotypic features consistent with a
marginalzone origin: is this a distinct entity?. Blood. 2014;123(8):
11991206.
9. Bruscaggin A, Monti S, Arcaini L, et al. Molecular lesions of signalling
pathway genes in clonal Bcell lymphocytosis with marginal zone
features. Br J Haematol. 2014;167(5):718720.
10. Mamessier E, BroussaisGuillaumot F, Chetaille B, et al. Nature and
importance of follicular lymphoma precursors. Haematologica.
2014;99(5):802810.
11. Schmidt J, Salaverria I, Haake A, et al. Increasing genomic and
epigenomic complexity in the clonal evolution from in situ to
manifest t(14;18)positive follicular lymphoma. Leukemia. 2014;
28(5):11031112.
12. Schmidt J, RamisZaldivar JE, Bonzheim I, et al. CREBBP gene mu-
tations are frequently detected in in situ follicular neoplasia. Blood.
2018;132(25):26872690.
13. Hellmuth JC, Louissaint A, Jr., Szczepanowski M, et al. Duodenal
type and nodal follicular lymphomas differ by their immune micro-
environment rather than their mutation profiles. Blood. 2018;
132(16):16951702.
14. Schmatz AI, Streubel B, KretschmerChott E, et al. Primary follicular
lymphoma of the duodenum is a distinct mucosal/submucosal
variant of follicular lymphoma: a retrospective study of 63 cases.
J Clin Oncol. 2011;29(11):14451451.
15. Louissaint A, Jr., Schafernak KT, Geyer JT, et al. Pediatrictype nodal
follicular lymphoma: a biologically distinct lymphoma with frequent
MAPK pathway mutations. Blood. 2016;128(8):10931100.
16. Liu Q, Salaverria I, Pittaluga S, et al. Follicular lymphomas in children
and young adults: a comparison of the pediatric variant with usual
follicular lymphoma. Am J Surg Pathol. 2013;37(3):333343.
17. Schmidt J, Gong S, Marafioti T, et al. Genomewide analysis of
pediatrictype follicular lymphoma reveals low genetic complexity
and recurrent alterations of TNFRSF14 gene. Blood. 2016;128(8):
11011111.
18. Schmidt J, RamisZaldivar JE, Nadeu F, et al. Mutations of MAP2K1
are frequent in pediatrictype follicular lymphoma and result in ERK
pathway activation. Blood. 2017;130(3):323327.
19. Woessmann W, QuintanillaMartinez L. Rare mature Bcell lym-
phomas in children and adolescents. Hematol Oncol. 2019;37(Suppl
1):5361.
How to cite this article: QuintanillaMartinez L. Recognizing
but not harming. Borderline Bcell lymphoid proliferations.
Hematological Oncology. 2021;39(S1):6167. https://doi.org/
10.1002/hon.2854
QUINTANILLAMARTINEZ
-
67
Received: 15 February 2021
DOI: 10.1002/hon.2848
SUPPLEMENT ARTICLE
Sequencing of myeloma therapy: Finding the right path
among many standards
S. Vincent Rajkumar
Division of Hematology, Mayo Clinic, Rochester, Minnesota, USA
Correspondence
S. Vincent Rajkumar, Division of Hematology, Mayo Clinic, 200 First Street SW, Rochester, MN 55905, USA.
Email: rajkumar.vincent@mayo.edu
In the last decade, major changes have occurred in the diagnostic
criteria, staging system, and response criteria for multiple myeloma
(MM).
1,2
These changes have been accompanied by several advances
in treatment of the MM, including many new drugs (carfilzomib,
pomalidomide, daratumumab, elotuzumab, panobinostat, ixazomib,
selinexor, isatuximab, and belantamab). Numerous clinical trials
provide data on best practices along the spectrum of the disease. The
purpose of this article is to describe the complexity of MM therapy in
a landscape where many parallel standards exist for the same
indications, and to provide an outline selecting the first line and
subsequent sequencing of therapy according to dynamic changes in
the disease and patient features.
1
|
INTIAL THERAPY
Approach to initial therapy in myeloma is affected by host factors
(age, performance status, renal function), eligibility for stem cell
transplantation, and presence or absence of high risk features. High
risk MM is defined by the presence of t(4;14), t(14;16), t(14;20),
deletion 17p, gain 1q, or p53 mutation.
3
Doublehit MM refers to the
presence of any two or more highrisk abnormalities. Triplehit MM
refers to the presence of three or more highrisk abnormalities.
Although minimal residual disease (MRD) negative status is associ-
ated with improved progressionfree survival (PFS) and overall
survival (OS), there are no data from randomized trials that
modifying therapy in MRD positive patients in an attempt to make
them MRD negative will lead to better outcomes.
The current algorithms for the treatment of symptomatic newly
diagnosed MM is shown in Figure 1. Patients who candidates for
autologous stem cell transplantation (ASCT) are treated with three to
four cycles of induction therapy followed by stem cell harvest. After
stem cell harvest, the standard of care has been ASCT followed by
maintenance. However, in selected patients who have standard risk
MM, ASCT can be delayed until relapse, and randomized trials show
no detrimental effect on OS with such an approach.
The preferred initial therapy is bortezomib, lenalidomide, and
dexamethasone (VRd).
4
The 4year OS rate with VRd is greater than
80% with or without early ASCT.
5
An important alternative to VRd
in newly diagnosed MM is daratumumab, lenalidomide, and
dexamethasone (DRd). DRd has shown significant efficacy in a
randomized trial conducted in transplant ineligible patients, with
improved PFS compared with Rd.
6
However, there are important
differences between the two approaches, particularly in transplant
ineligible patients where VRd consists of triplet therapy for 6 months
followed by maintenance, while DRd requires continuous triplet
therapy until disease progression. Thus due to overall cost, and
strength of longterm data, I prefer VRd over DRd for most patients.
7
However, DRd is a suitable alternative for patients with preexisting
neuropathy or for patients who have intolerance to VRd. In highrisk
transplant eligible patients with double hit MM or triple hit MM,
I recommend addition of daratumumab to the standard VRd regimen
(DaraVRd), or the daratumumab, bortezomib, thalidomide, and
dexamethasone (DaraVTd) regimen.
8
There is no significant benefit
with carfilzomib, lenalidomide, and dexamethasone (KRd) over VRd in
newly diagnosed MM.
9
KRd is more expensive, and is associated with
a higher risk of serious cardiac, renal, and pulmonary toxicity
than VRd.
In certain settings, the treatment regimens for newly diagnosed
MM must be modified. Thus, bortezomib, cyclophosphamide, and
dexamethasone (VCd) is the preferred regimen in patients presenting
with acute renal failure due to light chain cast nephropathy. In
patients with primary plasmacell leukemia multiagent combination
chemotherapy such as bortezomib/dexamethasone/thalidomide/
cisplatin/doxorubicin/cyclophosphamide/etoposide (VDTPACE) is
usually needed initially to achieve rapid disease control. In elderly
frail patients who are unable to travel to receive parenteral therapy,
the all oral regimen of ixazomib, lenalidomide, and dexamethasone
68
-
Hematological Oncology. 2021;39(S1):6872. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
(IxaRd) is a reasonable alternative to VRd and DRd. Melphalan
based regimens are no longer recommended for newly diagnosed
MM due to concerns about stem cell damage, secondary myelodys-
plastic syndrome, and acute leukemia.
1.1
|
Timing of stem cell transplantation
One of the main controversies in initial therapy is concerning the
sequencing of ASCT. Older randomized trials have found similar OS
with early ASCT (immediately following four cycles of induction
therapy) versus delayed ASCT (at the time of relapse as salvage
therapy). More recently, the Intergroupe Francophone du Myelome
(IFM) trial found no significant OS difference between early versus
delayed ASCT in patients treated with VRd as initial therapy and
lenalidomide maintenance.
10
No difference has emerged even after
8 years of followup. There is a significant improvement in PFS as
expected with early ASCT, and there are other logistical benefits to
ASCT. In general, early ASCT is preferred, but based on the IFM
results it is reasonable to consider a delayed ASCT in patients with
standardrisk multiple MM who prefer such an approach for personal
reasons.
The role of consolidation therapy and tandem (double) ASCT
is limited. Results of randomized trials are contradictory and
likely reflect the availability of new treatment options in the
salvage setting. In the United States, where multiple options for
salvage therapy are available, there seems to be no benefit with
tandem ASCT. At present, outside of a clinical trial setting, we
consider tandem ASCT only in selected young patients with
del 17p.
FIGURE 1 Approach to the treatment of newly diagnosed myeloma in transplant eligible (A) and transplant ineligible (B) patients. ASCT,
autologous stem cell transplantation; DaraVRd, daratumumab, bortezomib, lenalidomide, dexamethasone; DRd, daratumumab, lenalidomide,
dexamethasone; VRd, bortezomib, lenalidomide, dexamethasone. Modified from Rajkumar and Kumar
2
RAJKUMAR
-
69
1.2
|
Maintenance therapy
The standard of care following initial therapy MM is lenalidomide
maintenance. In highrisk patients, bortezomib plus lenalidomide, or
VRd maintenance is preferable. There are limited data on optimal
duration of maintenance. Many patients can benefit from a drugfree
interval, and trials are examining if the duration of maintenance can
be modified based on MRD results.
1.3
|
Relapsed MM
MM is characterized by multiple remissions and relapses. The
sequencing of therapies appropriately is critical in achieving the best
longterm survival. Many patients with MM receive five or more
lines of therapy in a sequential manner over several years. The
choice of treatment at each relapse is affected by the Timing of the
relapse, Response to prior therapy, Aggressiveness of the relapse,
and Performance status. Patients are eligible for ASCT should be
considered for transplantation if they had elected to delay the
procedure, or if they achieved excellent remission duration with the
first ASCT, defined as a remission of 36 months or longer with
maintenance.
In order to provide the best sequential therapy, we need to
select the most active regimen early on. A triplet regimen is preferred
at first relapse. At each subsequent relapse, a triplet, quadruplet, or
multidrug regimen that contains at least two new drugs that the
patient is not refractory to should be used. The algorithm for the
treatment of relapsed MM is given in Figure 2.
Treatment is typically continued until disease progression.
However, based on tolerability and response, increasing the interval
between cycles, as well as treatmentfree intervals should be
considered.
At first relapse, my preferred option is DRd for patients who are
not refractory to lenalidomide. If patients are refractory to lenali-
domide, the choices are daratumumab, bortezomib dexamethasone
FIGURE 2 Approach to the treatment of relapsed multiple myeloma in first relapse (A) and second or higher relapse (B). DKd,
daratumumab, carfilzomib, and dexamethasone; DPd, daratumumab, pomalidomide, and dexamethasone; DRd, daratumumab, lenalidomide,
and dexamethasone; DVd, daratumumab, bortezomib, and dexamethasone; ERd, elotuzumab, lenalidomide, and dexamethasone; IPd, ixazomib,
pomalidomide, and dexamethasone; IRd, ixazomib, lenalidomide, and dexamethasone; KPd, carfilzomib, pomalidomide, and dexamethasone;
KRd, carfilozomib, lenalidomide, and dexamethasone; VCd, bortezomib, cyclophosphamide. Modified from Rajkumar and Kumar
2
70
-
RAJKUMAR
(DVd), daratumumab pomalidomide, dexamethasone, or isatuximab,
pomalidomide, dexamethasone. In patients who are refractory to
daratumumab at first relapse, carfilzomibbased regimens such as
KRd or carfilzomib, pomalidomide, dexamethasone (KPd) are excel-
lent options. For patients who are frail, IxaRd would be a reasonable
first choice for relapse.
There are numerous other alternatives, and these can be used in
second and subsequent relapses. They include elotuzumab, pomali-
domide, dexamethasone (EPd), bortezomib, cyclophosphamide,
dexamethasone (VCd), bortezomib, pomalidomide, dexamethasone
(VPd), and daratumumab, carfilzomib, and dexamethasone (DKd).
Unfortunately, none of the triplet regimens used in relapsed MM
have been compared headtohead in randomized trials.
When sequencing therapy, there are a few other important
considerations. At each relapse, any of the regimens that were
mentioned for use in first relapse can be considered, with the goal of
having at least two new drugs that the patient is not refractory two,
and preferably from a different drug class. In many instances, this
may mean the necessity of adding a monoclonal antibody to one of
the triplets to create a quadruplet regimen. Although it is in the same
drug class, pomalidomide has clinical activity in patients who are
refractory to lenalidomide. Similarly, carfilzomib has activity in
patients who are refractory to bortezomib. Carfilzomib is typically
administered twiceweekly at a dose of 27 mg/m
2
, but a onceweekly
schedule of 56–70 mg/m
2
may be equally effective and safe, and
more convenient.
11
Carfilzomib has a lower risk of neurotoxicity than
bortezomib, but approximately 5% of patients can experience serious
cardiac side effects.
1.4
|
Refractory myeloma
There are several additional options for patients with MM
refractory to immunomodulatory drugs, proteasome inhibitors,
alkylators, CD38 antibodies, and elotuzumab. One option is to add
panobinostat to a proteasomeinhibitor containing regimen. A
second option is to use a selinexorcontaining regimen such as
selinexor, bortezomib, dexamethasone (SVd). A third options is
treatment with belantamab mafodotin, a humanized antiBCMA
antibody that is conjugated to monomethyl auristatinF, a micro-
tubule disrupting agent.
12
Other options for refractory disease
include bendamustinecontaining regimens or anthracycline
containing regimens.
Venetoclax is not approved for use in MM, but has singleagent
activity in patients with t(11;14) subtype of MM.
13
A recent ran-
domized trial found significantly higher mortality with venetoclax in
relapsed MM despite producing deeper responses and better PFS.
14
Therefore, venetoclax is considered investigational, and its use
should be restricted to patients with t(11;14) who have relapsed
disease.
With careful analysis of the various options and combinations
possible, we can induce remissions multiple times with creative
strategies. At each step opportunities for clinical trials may open up
and should be considered.
1.5
|
Investigational treatments
One of the most exciting options being studied is chimeric antigen
receptor T cells targeting B cell maturation antigen such as bb2121.
15
In studies so far, more than 80% of patients appear to respond, with
median response duration of approximately 12 months. Another
promising new strategy is the use of bispecific Tcell engager, such as
AMG 701, talquetamab, or cevostamab. Iberdomide and other
cereblon inhibitors are also showing promise.
ACKNOWLEDGMENTS
This study was supported in part by grants CA 107476, CA 168762,
and CA186781 from the National Cancer Institute, Rockville, MD,
USA.
CONFLICT OF INTERESTS
The author declares that there is no conflict of interest.
REFERENCES
1. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International
Myeloma Working Group updated criteria for the diagnosis of
multiple myeloma. Lancet Oncol. 2014;15:e538e548.
2. Rajkumar SV, Kumar S. Multiple myeloma current treatment
algorithms. Blood Cancer J. 2020;10:94.
3. Rajkumar SV. Multiple myeloma: 2020 update on diagnosis,
riskstratification and management. Am J Hematol. 2020;95:
548567.
4. Durie BGM, Hoering A, Abidi MH, et al. Bortezomib, lenalido-
mide and dexamethasone vs. Lenalidomide and dexamethasone
induction followed by lenalidomide and dexamethasone mainte-
nance in patients with newly diagnosed myeloma without intent
for immediate autologous stem cell transplant: results of the
randomised phase III SWOG trial S0777. Lancet. 2017;389:
519527.
5. Attal M, LauwersCances V, Hulin C, et al. Autologous trans-
plantation for multiple myeloma in the era of new drugs: a phase III
study of the Intergroupe Francophone du Myelome (IFM/DFCI 2009
trial). Blood. 2015;126:391.
6. Facon T, Kumar S, Plesner T, et al. Daratumumab plus lenalidomide
and dexamethasone for untreated myeloma. N Engl J Med. 2019;380:
21042115.
7. Kapoor P, Rajkumar SV. MAIA under the microscope—
bringing trial design into focus. Nat Rev Clin Oncol. 2019;16:
339340.
8. Moreau P, Attal M, Hulin C, et al. Bortezomib, thalidomide, and
dexamethasone with or without daratumumab before and after
autologous stemcell transplantation for newly diagnosed multiple
myeloma (CASSIOPEIA): a randomised, openlabel, phase 3 study.
Lancet. 2019;394:2938.
9. Kumar SK, Jacobus SJ, Cohen AD, et al. Carfilzomib or bortezomib
in combination with lenalidomide and dexamethasone for newly
diagnosed myeloma without intent for immediate autologous
stemcell transplant (E1A11): a multicenter, open label, phase
3, randomised, controlled trial. Lancet Oncol. 2020;21(10):
13171330.
RAJKUMAR
-
71
10. Attal M, LauwersCances V, Hulin C, et al. Lenalidomide, bortezomib,
and dexamethasone with transplantation for myeloma. N Engl J Med.
2017;376:13111320.
11. Bringhen S, Mina R, Petrucci MT, et al. Onceweekly versus twice
weekly carfilzomib in patients with newly diagnosed multiple
myeloma: a pooled analysis of two phase I/II studies. Haematologica.
2019;104:16401647.
12. Lonial S, Lee HC, Badros A, et al. Belantamab mafodotin for relapsed
or refractory multiple myeloma (DREAMM2): a twoarm,
randomised, openlabel, phase 2 study. Lancet Oncol. 2020;21:
207221.
13. Kumar S, Kaufman JL, Gasparetto C, et al. Efficacy of venetoclax as
targeted therapy for relapsed/refractory t(11;14) multiple myeloma.
Blood. 2017;130:24012409.
14. Kumar S, Rajkumar SV. Surrogate endpoints in randomised
controlled trials: a reality check. Lancet. 2019;394:281283.
15. Raje N, Berdeja J, Lin Y, et al. AntiBCMA CAR Tcell therapy
bb2121 in relapsed or refractory multiple myeloma. N Engl J Med.
2019;380:17261737.
How to cite this article: Rajkumar SV. Sequencing of myeloma
therapy: Finding the right path among many standards.
Hematological Oncology. 2021;39(S1):6872. https://doi.org/
10.1002/hon.2848
72
-
RAJKUMAR
Received: 15 February 2021
DOI: 10.1002/hon.2849
SUPPLEMENT ARTICLE
Molecular diagnostics and reporting in lymphoid
malignancies: Current status and beyond
Richard Rosenquist
Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm, Sweden
Correspondence
Richard Rosenquist, Department of Molecular Medicine and Surgery, Karolinska Institutet, Stockholm 17176, Sweden.
Email: richard.rosenquist@ki.se
Funding information
The Swedish Cancer Society; The Swedish Research Council; The Knut and Alice Wallenberg Foundation; Radiumhemmets Forskningsfonder, Stockholm
1
|
INTRODUCTION
Thanks to the advent of nextgeneration sequencing (NGS) technol-
ogies more than 10 years ago, the genomic landscape of most larger
cancer types has been unraveled. Hematological and lymphoid ma-
lignancies were among the first cancer types to be sequenced,
1
probably as they are more easily accessible. Based on these studies,
the number of clinically relevant genetic aberrations with diagnostic,
prognostic, and/or predictive impact has increased rapidly.
2
This
increases demands on molecular diagnostics to ensure that different
types of genetic alterations can be readily detected in a diagnostic
setting and within a reasonable time frame.
For hematological malignancies, genetics have been an integral
part of diagnostics since decades,
3
including methods such as cyto-
genetics, fluorescence in situ hybridization (FISH) and targeted
mutational analysis, and/or Sanger sequencing (Figure 1). With the
more powerful NGS technologies, targeted gene panels have been
added to the arsenal of methods in the diagnostic laboratory.
Regardless of the type of diagnostic test, it is essential to harmonize
methodology according to international standards and guidelines. To
ensure comparable results, especially in multicenter studies, it is
equally important that the interpretation of the results and the
clinical reporting follow established guidelines. This is particularly
relevant as genetic test results are becoming more complex with
different NGS assays covering a range of different genetic
aberrations.
This paper gives an overview of different stateoftheart molec-
ular technologies that are applied in clinical diagnostics of lymphoid
malignancies, their advantages, and limitations. It also discusses
aspects that need to be considered to harmonize clinical interpre-
tation and reporting of NGS data.
2
|
TECHNOLOGIES APPLIED FOR DIAGNOSTICS
OF LYMPHOID MALIGNANCIES
2.1
|
Molecular cytogenetics
Cytogenetics, or chromosome banding analysis, has long been
instrumental to define recurrent chromosomal aberrations observed
in different lymphoid malignancies. While cytogenetics has remained
central in leukemias (i.e., acute myeloid leukemia [AML], acute
lymphoblastic leukemia [ALL], and chronic myeloid leukemia),
molecular cytogenetics or FISH is preferred for mature lymphoid
malignancies. The introduction of the FISH technology enabled
analysis not only on metaphase chromosomes, but also on interphase
chromosomes (i.e., nondividing cells). FISH is therefore more rapid
than cytogenetics and can be used to screen for recurrent diagnostic
or risk stratifying genomic aberrations. Moreover, FISH can be
applied to various tissue types, such as blood smears, imprints as well
as formalinfixed, paraffinembedded (FFPE) tissue. Robust, com-
mercial FISH probes are available for most recurrent aberrations,
although they are relatively expensive. It is commonly used to detect
translocations in lymphomas; for example, t(11;14) in mantle cell
lymphoma and MYC and BCL2 translocations in diffuse large Bcell
lymphoma (DLBCL).
3
In chronic lymphocytic leukemia (CLL), a panel
of probes is usually applied to detect riskstratifying aberrations, that
is, 11qdeletion, 13qdeletion, 17pdeletion, and trisomy 12, whereas
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, pro-
vided the original work is properly cited.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
Hematological Oncology. 2021;39(S1):7377. wileyonlinelibrary.com/journal/hon
-
73
in multiple myeloma, probes are used to identify t(4;14), t(11;14),
t(14;16), 1q gain, and 17p deletion. However, while FISH on inter-
phase cells can give a quick answer as to whether a specific genetic
aberration is present or not, it is not a particularly sensitive
technique; the cutoff level to detect an aberration has to be deter-
mined by each laboratory and is usually around 5%.
2.2
|
Targeted mutation analysis
Before the NGS era, most clinical laboratories applied Sanger
sequencing for targeted mutation analysis. Sanger sequencing has
been the gold standard since the 1980s and is a very robust
technology. However, it rapidly becomes expensive and time
consuming if several genes or large genes are investigated. Another
limitation with Sanger sequencing is its sensitivity which is usually
between 10% and 20%.
In CLL, TP53 mutations are linked to poor response to chemo-
immunotherapy and poor overall survival, and TP53 gene screening is
nowadays mandatory before start of any line of treatment.
4
TP53
gene analysis (encompassing exons 211) has typically been
performed by Sanger sequencing, although more recently many
laboratories have shifted to NGS.
Another sequencedbased molecular test is the immunoglobulin
heavy variable (IGHV) gene mutational status that defines CLL with
unmutated IGHV genes and an inferior outcome, and CLL with
mutated IGHV genes and an expected favorable prognosis. This
analysis is mainly performed by polymerase chain reaction (PCR)
amplification of the clonal IGH rearrangement followed by Sanger
sequencing,
5
although new NGS protocols have been developed.
6
For both TP53 analysis and IGHV gene analysis, the European
Research Initiative on CLL (ERIC) has provided technical recom-
mendations and guidance for interpretation.
4,5
It has also imple-
mented dedicated certification systems that enable clinical
laboratories to regularly certify their method (Sanger sequencing or
NGS) against standard operation procedures.
For hotspot mutation detection, such as BRAF (V600E) in hairy
cell leukemia and MYD88 mutations (L265P) in Waldenström's
macroglobulinemia, specific assays have been established. These use
allelespecific PCR or quantitative PCR to detect the mutant allele.
More recently, digital droplet PCR has been developed that enables a
very sensitive detection (down to 0.01%) and can be applied to detect
recurrent mutations or to follow patients over time. It is, for instance,
used to detect BTK/PLCG2 mutations in CLL patients progressing on
ibrutinib treatment.
2.3
|
Nextgeneration sequencing
Since more than 10 years, we have had access to new types of
sequencing instruments that provide massive parallel sequencing or
NGS. Using this technology, we can analyze the entire genome
(wholegenome sequencing [WGS]), the exome (wholeexome
sequencing), or selected regions of particular interest, that is,
targeted NGS or gene panels. Depending on the detail of sequencing,
the number of obtained sequence reads differs. Using WGS, the
recommendation for tumor samples is to sequence to a sequence
depth of 90(and 30for the matched normal sample). For targeted
gene panels, depending on the size of the panel, one aims for a high
sequence depth of at least 500but preferably above 1000.
Until recently, the majority of gene panels used within clinical
diagnostics were ampliconbased, meaning they use PCR to amplify
selected amplicons covering the genes/exons of interest. These
ampliconbased panels can be implemented relatively easily using
either a panel that is designed inhouse, or commercially available
panels. However, the limitations with ampliconbased technology are
potential biases in amplification, particularly in difficult to sequence
regions (i.e., GC rich and/or repetitive regions), that can cause
dropout of amplicons, as well as an increased risk of lowfrequency,
falsepositive mutations; this can happen in particular when FFPE
material is used as source of input.
In an attempt to study the robustness of ampliconbased gene
panels, we within ERIC recently conducted a multicenter study using
three different technologies (Multiplicom, HaloPlex, and TruSeq),
targeting 11 genes recurrently mutated in CLL.
7
All six centers
analyzed the same 48 CLL samples, and each technique was analyzed
FIGURE 1 Evolution of genetic diagnostics. FISH, fluorescence in situ hybridization
74
-
ROSENQUIST
by two centers. A very high concordance was achieved between
technologies and centers for gene mutations with a variant allele
frequency (VAF) above 5%, while discrepancies started to appear for
variants with a VAF <5%. Hence, we conclude that ampliconbased
sequencing can be safely adopted for somatic mutation detection
with VAFs >5%.
We also tested a highsensitivity assay containing unique
molecular identifiers (UMIs) which could confirm subclonal mutations
with a VAF <5%.
7
In this approach, duplicate reads are removed
which improved sensitivity. Therefore, inclusion of UMIs should be
considered when new panels are designed.
More recently, capturebased enrichment panels, using baits
(probes) to hybridize to the region of interest, have been developed.
These panels are usually larger in size (hundreds of genes) and
produce more even sequence reads. This means that more “prob-
lematic” regions can be sequenced. Another advantage is that they
enable simultaneous detection of different types of genomic
aberrations (as they are larger in size). For instance, in addition to
investigating a selected number of genes for single nucleotide
variants (SNVs) and insertions/deletions (indels), it is also possible to
analyze copynumber changes (i.e., deletions and amplifications) and
structural variants (e.g., translocations). He et al.
8
combined a DNA
and RNAbased broad, capturebased gene panel for diagnostics of a
large number of hematological malignances (n=3696) including
different lymphoid malignancies. Using either bone marrow or FFPE
samples, they could detect all aforementioned types of genomic
aberrations with high accuracy and reproducibility.
Gene panels have rapidly been introduced into routine
diagnostics of myeloid malignancies. In Sweden, we recently shifted
from a 54 gene ampliconbased gene panel to a national capture
based myeloid panel including 195 genes. In lymphoid malignancies,
smaller ampliconbased gene panels are currently in clinical use. As
mentioned, in CLL, we test for TP53 mutations before start of
treatment and many centers have switched to ampliconbased
NGSgene panels. There are also additional genes of diagnostic,
prognostic, and predictive impact in lymphomas that can be captured
with these smaller panels. For instance, detection of certain muta-
tions has diagnostic utility, for example, MYD88 mutations in
Waldenström's macroglobulinemia, BRAF mutations in hairy cell
leukemia, KLF2 mutations in splenic marginal zone lymphomas, and
STAT3/STAT5B mutations in Tcell lymphomas, while other gene
mutations are associated with response to treatment (e.g., MYD88/
CXCR4 mutations in Waldenström's macroglobulinemia patients
treated with ibrutinib) or treatment resistance (e.g., BTK/PLCG2
mutations in ibrutinibtreated CLL).
2
Considering that a broad spectrum of genetic aberrations is
involved in ontology and evolution of lymphoid malignancies,
1
the
introduction of capturebased panels will be particularly useful in this
patient group to capture not only SNV/indels, but also copynumber
aberrations (CNAs) and translocations
9
as well as more complex
markers such as IG/Tcell receptor gene analysis. In Sweden, we have
recently developed a (national) capturebased lymphoid panel
including 252 genes that is currently under validation and will soon
be implemented into routine diagnostics.
3
|
REPORTING AND INTERPRETATION OF
RESULTS
For all types of molecular reports, it is important to include a number
of key parameters so that the results can be easily understood by
other laboratories. The report should include general information,
such as personal id, referring doctor, and more specific information,
such as tissue type investigated, method applied, and genetic aber-
rations assessed (Table 1). For FISH analysis, it is important to state
the probes used and number of cells investigated as well as the cutoff
applied. For more targeted analyses, such as hotspot mutation
detection, the technology, and sensitivity should also be provided.
For NGSbased analysis, it is important to state gene coverage
and sequence depth as well as cutoffs for variant calling. The
sequence variants should be listed and include variant description at
cDNA/protein level, following the HGVS nomenclature, the number
of variant reads versus total number of sequence reads and the VAF
(Table 1). It should also be noted if the variant has been deemed as
pathogenic using locusspecific databases or international guidelines
(e.g., following the American College of Medical Genetics (ACMG)
criteria
10
). Furthermore, in the concluding remark of the report, it
should be stated if a somatic variant has been detected before in this
disease entity that has diagnostic, prognostic, or predictive impact,
based on the WHO classification,
3
clinical consensus guidelines, or
available published literature. More complex data that includes CNAs
and/or structural aberrations will require continued development of
bioinformatics tools to visualize the reported data, for example, in
the format of copynumber plots or circus plots.
As we enter the precision medicine era, more and more targeted
therapies have become available in oncology. Some academic centers
and commercial companies have developed new support systems for
clinical decisionmaking that assist in defining if a mutation is
considered “actionable” or not.
11
Based on large databases, these
tools can, for a certain variant, provide an alert if a clinically relevant
finding has been made and provide links to available potential
targeted drugs and/or clinical trials (e.g., FDAapproved drugs,
ongoing clinical trials, or if a drug has been used for another malig-
nancy). Thus far, these support systems have been mainly used for
solid cancer and less frequently in hematological malignancies.
4
|
CONCLUDING REMARKS
Genetic diagnostics within hematological malignancies has over the
years evolved dramatically from cytogenetics to targeted NGS
strategies (Figure 1). For lymphoid malignancies, we apply different
FISH analysis as well as targeted sequencing/NGS for diagnostic and
riskstratifying purposes.
3
However, with the everincreasing
ROSENQUIST
-
75
TABLE 1Data to be included in the molecular report based on sequencing
Basic data: Patient data (name, date of birth and/or id number)
Diagnosis
Type of material (peripheral blood, bone marrow, lymph node)
Date of sample collection/delivery
Requested by (department/hospital)
Method: Sequencing strategy: Sanger sequencing, ampliconbased NGS or capturebased NGS
Genes/exons analyzed, minimum/mean sequence depth, detection limit, reference
sequence
Results: Variant description (cDNA/protein level)
Variant type (missense/nonsense/frameshift)
Variant allele frequency (number of variant reads/total number of reads)
Variant pathogenicity
Conclusion: Clinical interpretation of the variant detected and summary of its clinical impact
(diagnostic, prognostic and/or predictive) according to current knowledge
Abbreviation: NGS, nextgeneration sequencing.
number of clinically relevant genetic variants that are detected in
lymphoid malignancies, we need to continue developing “dynamic”
NGSbased strategies, such a capturebased sequencing, that include
large numbers of genes and types of genetic aberrations. In this
regard, a recent study identified seven novel distinct subgroups of
DLBCL based on broad genetic characterization.
12
This illustrates
the need to develop more comprehensive analyses for diagnostic
purposes, also to carry out future precision medicine studies.
Furthermore, it is important to develop decision support tools that
can inform clinicians on “actionability” for the different genetic
events detected.
Within rare disease diagnostics, WGS has increasingly replaced
multigene testing.
13
Could this be a way forward also for hemato-
logical malignancies where we usually perform multiple testing? Some
national and/or regional programs are now testing if WGS combined
with RNAsequencing could replace the “old” technologies in acute
leukemia (ALL and AML). In a proofofconcept study, Klintman et al.
demonstrated that WGS and targeted NGS had a high concordance
for SNVs/indels in CLL, while the concordance between FISH and
WGS was lower.
14
Hence, before we can start to use wholegenome
techniques, we have to be certain that all mandatory genomic aber-
rations can be detected and provided within a reasonable time frame.
Another rapidly evolving areas of interest are liquid biopsies and
the detection of circulating tumor DNA, which have the potential to
enable sensitive NGSbased followup of lymphoma patients and can
detect genetic aberrations in cases where it is difficult to take a
biopsy.
15
These analyses have not yet entered clinical use in
lymphoid malignancies, but they are foreseen to be an important part
of future diagnostics.
Finally, although there are a few ongoing precision medicine
studies in lymphoid malignancies worldwide, it will be very impor-
tant to initiate future clinical trials based on targeted drugs/immune
therapy to realize the full potential of precision medicine.
CONFLICT OF INTEREST
Richard Rosenquist has received honoraria from Abbvie, AstraZe-
neca, Illumina, Janssen and Roche.
PEER REVIEW
The peer review history for this article is available at https://publons.
com/publon/10.1002/hon.2849
ORCID
Richard Rosenquist https://orcid.org/0000-0002-0211-8788
REFERENCES
1. Rosenquist R, Bea S, Du MQ, Nadel B, PanHammarstrom Q. Genetic
landscape and deregulated pathways in Bcell lymphoid malig-
nancies. J Intern Med. 2017;282(5):371394.
2. Rosenquist R, Rosenwald A, Du MQ, et al. Clinical impact of recur-
rently mutated genes on lymphoma diagnostics: stateoftheart and
beyond. Haematologica. 2016;101(9):10021009.
3. Swerdlow SHCE, Harris NL, Jaffe ES, Pileri SA, Stein H, Thiele J.
WHO Classification of Tumours of Haematopoietic and Lymphoid
Tissues. Rev. 4th ed. 2017:586.
4. Malcikova J, Tausch E, Rossi D, et al. ERIC recommendations for
TP53 mutation analysis in chronic lymphocytic leukemiaupdate on
methodological approaches and results interpretation. Leukemia.
2018;32(5):10701080.
5. Rosenquist R, Ghia P, Hadzidimitriou A, et al. Immunoglobulin gene
sequence analysis in chronic lymphocytic leukemia: updated ERIC
recommendations. Leukemia. 2017;31(7):14771481.
6. Davi F, Langerak AW, de Septenville AL, et al. Immunoglobulin gene
analysis in chronic lymphocytic leukemia in the era of next genera-
tion sequencing. Leukemia. 2020;34(10):25452551.
7. Sutton LA, Ljungstrom V, Enjuanes A, et al. Comparative analysis of
targeted nextgeneration sequencing panels for the detection of
gene mutations in chronic lymphocytic leukemia: an ERIC multi
center study. Haematologica. 2020.
8. He J, AbdelWahab O, Nahas MK, et al. Integrated genomic DNA/
RNA profiling of hematologic malignancies in the clinical setting.
Blood. 2016;127(24):30043014.
76
-
ROSENQUIST
9. Wren D, Walker BA, Bruggemann M, et al. Comprehensive trans-
location and clonality detection in lymphoproliferative disorders by
nextgeneration sequencing. Haematologica. 2017;102(2):e57e60.
10. Richards S, Aziz N, Bale S, et al. Standards and guidelines for the
interpretation of sequence variants: a joint consensus recommen-
dation of the American College of Medical Genetics and Genomics
and the Association for Molecular Pathology. Genet Med. 2015;17(5):
405424.
11. Tamborero D, Dienstmann R, Rachid MH, et al. Support systems to
guide clinical decisionmaking in precision oncology: the Cancer
Core Europe Molecular Tumor Board Portal. Nat Med. 2020;26(7):
992994.
12. Wright GW, Huang DW, Phelan JD, et al. A probabilistic classifica-
tion tool for genetic subtypes of diffuse large B cell lymphoma with
therapeutic implications. Cancer Cell. 2020;37(4):551568.
13. Lindstrand A, Eisfeldt J, Pettersson M, et al. From cytogenetics to
cytogenomics: wholegenome sequencing as a firstline test
comprehensively captures the diverse spectrum of diseasecausing
genetic variation underlying intellectual disability. Genome Med
2019;11(1):68.
14. Klintman J, Barmpouti K, Knight SJL, et al. Clinicalgrade validation
of whole genome sequencing reveals robust detection of low
frequency variants and copy number alterations in CLL. Br J
Haematol. 2018;182(3):412417.
15. Rossi D, Diop F, Spaccarotella E, et al. Diffuse large Bcell lymphoma
genotyping on the liquid biopsy. Blood. 2017;129(14):19471957.
How to cite this article: Rosenquist R. Molecular diagnostics
and reporting in lymphoid malignancies: Current status and
beyond. Hematological Oncology. 2021;39(S1):7377. https://
doi.org/10.1002/hon.2849
ROSENQUIST
-
77
DOI: 10.1002/hon.2859
SUPPLEMENT ARTICLE
New drugs and pharmacological interactions in real life
Anastasios Stathis
New Drugs Development Unit, Oncology
Institute of Southern Switzerland, Bellinzona,
Switzerland
Correspondence
Anastasios Stathis, HeadNew Drugs
Development Unit, Oncology Institute of
Southern Switzerland, 6500 Bellinzona,
Switzerland.
Email: anastasios.stathis@eoc.ch
Abstract
A high number of new drugs have entered clinical development and many of them
have recently been approved for patients with lymphoid malignancies. The avail-
ability of new drugs offers additional treatment options, but it also requires
particular attention for the emergence of adverse events. In addition, new drugs
may also have interactions with other drugs, which could further increase the risk of
toxicities or result in decreased efficacy. Here we review potential drug interactions
for nonchemotherapy new drugs approved for patients with lymphoid malignancies.
KEYWORDS
drug interactions, lymphoid malignancies, new drugs
1
|
INTRODUCTION
Over the last years, several new drugs including small molecules,
monoclonal antibodies (either naked or conjugated), and more
recently adoptive cell therapies have been approved for the treat-
ment of patients with lymphoid malignancies.
1–8
With the exception
of chimeric antigen receptorT cells, most of the new drugs are
administered chronically (i.e., up to disease progression or relapse or
up to the occurrence of adverse events) aiming to achieve longterm
disease control. However, longterm drug administration can result in
adverse events and may also increase the risk of interactions with
other drugs.
While recognized as a potential risk, the frequency and severity
of drug–drug interactions (DDIs) in oncology is not clear. Here, we
review current knowledge regarding potential interactions that may
involve drugs that have become recently available for patients with
lymphoid malignancies and may affect every day clinical practice.
2
|
DDIs IN ONCOLOGY
Interactions among concomitantly administered drugs can result in
changes in the way one drug acts in the body and thus to altered
efficacy or toxicity. DDIs can derive from pharmacokinetic, pharma-
codynamic, or pharmaceutical interactions among two drugs (or
among a drug and alternative medications, herbs, or food).
9
A
pharmacokinetic interaction may affect any of the pharmacokinetic
properties (absorption, distribution, metabolism, and/or excretion
[ADME]) of one drug by another. The bestcharacterized is based on
cytochrome P450 (CYP) hepatic enzymes and occurs when drugs that
reduce (CYP inhibitors) or increase (CYP inducers) CYP activity are
concomitantly administered with CYP substrates resulting respec-
tively in decreases or increases in the metabolism of the substrate
drug. Interestingly, not only drugs but also food or herbs can have an
effect on CYP and therefore interfere with the metabolism of CYP
substrates, like grapefruit juice and Seville oranges that can act
respectively as strong or moderate CYP3A inhibitors and St. John's
wort that can induce CYP3A. Other pharmacokinetic interactions can
also occur, including interactions with the P glycoprotein 1 (Pgp)
drug transporter that can result in altered drug exposure and finally
altered pharmacokinetic properties.
10
Pharmacodynamic and pharmaceutical interactions occur,
respectively, when two drugs have similar mechanism of action (and
therefore can result in additive, synergistic, or antagonistic effects) or
when there are physical or chemical incompatibilities.
9
Drug interactions represent an important issue in oncology
given the older age of patients with cancer and the frequent use of
several medications (socalled polypharmacy) used to treat cancer
related symptoms or concomitant diseases.
11
Older series including
mainly patients with solid tumors reported that up to onethird of
cancer patients are exposed to potential DDIs.
12
Major potential
DDIs were identified in 16% of cancer patients in a large
78
-
Hematological Oncology. 2021;39(S1):7882. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
retrospective cohort
13
and up to 25% of patients on anticancer
treatments were found to have a potentially clinically significant
DDI in another study performed in one center.
14
A prospective trial
including also patients with hematological malignancies, reported
potential clinically relevant DDIs in 81 of 302 included patients
(27%).
15
Data from patients enrolled in phase II–IV clinical trials
with approved medications (mainly tyrosine kinases and monoclonal
antibodies) have been also published. DDIs that had to be avoided
or drugs to be used with caution were detected by protocol guid-
ance in 10% of patients, although the majority of subjects did not
have clinical relevant interactions based on pharmacist review. In
the same study, the use of the Lexicomp database detected mod-
erate to major DDIs in 24% of patients with 9.4% having a clinically
relevant DDI.
16
However, despite these data, the real frequency of DDIs in
oncology is unclear and there is lack of standardized criteria with
regards to clinical consequences and assessment of their severity.
17
For patients with lymphoid malignancies and especially those on
treatment with new drugs, data on DDIs are even scarcer. A recent
study performed in 118 patients with chronic lymphocytic leukemia
(CLL) on treatment with the Bruton's tyrosine kinase (BTK) inhibitor
ibrutinib showed that 64% of patients were on medications that
could increase ibrutinib toxicity and 3% on medications that could
decrease its efficacy.
18
Over the last years, improvements in our understanding of the
biology of lymphomas and advances in antibody technology have
permitted the development of many new compounds that have
become available for patients with lymphoma and CLL. These new
compounds comprise mainly small molecules and monoclonal anti-
bodies which have different mechanisms of action, different toxic-
ities, and potential for DDIs. Taking as a referral Food and Drug
Administration approvals of new drugs for lymphomas and CLL over
the last 5 years (Table 1), potential interactions, and recommenda-
tions for their management will be presented in the following
paragraphs.
3
|
BTK INHIBITORS
The development of BTK inhibitors has represented one of greatest
recent therapeutic achievements in the treatment of lymphoid
malignancies. Following the approval of ibrutinib, two other
compounds, acalabrutinib and zanubrutinib, have been recently
approved, while others are in clinical development and may be
added in the list of available therapies for the treatment of
lymphoid malignancies responding to BTK inhibitors currently
including CLL, mantlecell lymphoma, marginal lymphoma, and
Waldenström's macroglobulinemia.
BTK inhibitors represent a class of compounds with known
potential of pharmacokinetic DDIs. The firstinclass inhibitor ibru-
tinib is primarily metabolized by the cytochrome P450 CYP3A.
19
Although clinical trials in patients with CLL and lymphoma have
excluded concomitant use of strong CYP3A inhibitors or inducers,
pharmacokinetic studies in healthy volunteers and physiologically
based pharmacokinetic models have revealed changes in ibrutinib
exposure when administered concomitantly with CYP3A inhibitors or
inducers that may be clinically relevant.
20,21
Accordingly, it is rec-
ommended to avoid concomitant administration of ibrutinib with
strong CYP3A inhibitors or inducers and to consider a reduction of
its dose if a moderate CYP3A inhibitor must be used.
22
In addition to
the abovementioned CYP3Amediated interactions which may have a
clinical consequence, there is in vitro evidence that ibrutinib may also
interact with rituximab antagonizing its antibodydependent cell
mediated cytotoxicity though inhibition of interleukin2 inducible
tyrosine kinase which is necessary for natural killer cell function.
23
However, the clinical significance of this possible interaction is not
known.
Similarly to ibrutinib, the second generation BTK inhibitors
acalabrutinib and zunabrutinb are also primarily metabolized by
CYP3A, and they have the same recommendation of avoiding coad-
ministration with strong CYP3A inhibitors or inducers. Adaptations
of their dose should be considered when CYP3A moderate inducers
or inhibitors must be used.
24,25
In addition, acalabrutinib solubility
decreases with increasing gastric pH resulting in significant decreases
in exposer when administered with antacids and protonpump in-
hibitors. Therefore, the recommendation is that acalabrutinb should
not be coadministered with protonpump inhibitors (due to their
longlasting effect), while antacids and H2receptor antagonists may
be administered but at least 2 h after the administration of
acalabrutinib.
While the above reported interactions are based on pharmaco-
kinetic mechanism and modulation of CYP3A, there are other
potential interactions of BTK inhibitors that should be taken in
consideration. In particular, attention should be given to the
concomitant administration of anticoagulants which can lead to
increased risk of bleeding events. Concomitant administration of
warfarin is contraindicated. On the other hand, apixaban and rivar-
oxaban undergo CYP3A4mediated metabolism.
26,27
4
|
PHOSPHOINOSITIDE 3KINASE INHIBITORS
Another class of compounds that have entered clinical development
and have beshown activity mainly in CLL and some indolent lym-
phomas (follicular lymphoma in particular) is represented by phos-
phoinositide 3kinase (PI3K) inhibitors. Following the first approval
of idelalisib, other compounds targeting PI3K have been developed
more recently and two of them, copanlisib and duvelisib, have been
approved for follicular lymphoma.
Copanlisib is an intravenous, panclass I phosphatidylinositol3
kinase (PI3K inhibitor) with predominant PI3K‐α and PI3K‐δ inhibi-
tory activity. Approximately more than 90% of copanlisib metabolism
is mediated by CYP3A. Strong CYP3A inducers result in significant
decreases of copanlisib AUC and Cmax and should not be given
concomitantly. On the other hand, CYP3A strong inhibitors cause a
significant increase of copanlisib AUC and again should not be
STATHIS
-
79
administered concomitantly, or the dose of copanlisib should be
reduced in case concomitant use with strong inhibitors cannot be
avoided.
28
Finally, duvelisib, an inhibitor of PI3K with inhibitory activity
predominantly against PI3K‐δ and PI3K‐γ isoforms, is also primary
metabolized by CYP3A cytochrome and has the same indications as
with copanlisib for strong inducers or inhibitors. In addition, duvelisib
can lead to increase AUC of CYP3A substrates and therefore to
increased toxicity of these drugs which may require adaptation of
their dose.
29
5
|
VENETOCLAX
Another small moleculetargeted agent that has been approved for
the treatment of CLL is the bcl2 inhibitor venetcolax. As with the
compounds previously reported, venetoclax is mainly metabolized by
the cytochrome CYP3A4.
30
Concomitant use with a strong or mod-
erate CYP3A inhibitor or a Pgp inhibitor increases venetoclax
plasma concentration and exposure which may increase the risk of
adverse events, including tumor lysis syndrome, a wellknown
adverse event of venetoclax in CLL and reason for a particular
TABLE 1Selected recently approved drugs for lymphoid malignancies with known CYP3A and/or Pgp interactions. Refer to the
prescribing information of each drug.
Drug Recommendation Effects on other drugs
Ibrutinib Avoid concomitant use with strong CYP3A inhibitors. If these
inhibitors are used as shortterm consider interrupting
ibrutinib
May increase the concentration of oral Pgp or BCRP substrates
with narrow therapeutic index (e.g. digoxin, methotrexte).
Dose adjustments (70 mg OD or 140 mg OD) if voriconazole or
posaconazole must be given concomitantly
Dose adjustment to 280 mg OD if a moderate inhibitor must be
administered concomitantly
Strong CYP3A inducers should be avoided
Acalabrutinib Coadministration with strong CYP3A inhibitors should be avoided
or, if the inhibitor will be used shortterm, acalabrutinib should
be interrupted
Not reported
Dose adjustment at 100 mg OD if moderate CYP3A inhibitors are
used
Coadministration with strong CYP3A inducers should be avoided.
If they must be used consider increasing the dose of
acalabrutinib at 200 mg BID
Protonpump inhibitors should be avoided
Antacids and H2receptor antagonists to be taken at least 2 h after
acalabrutnib
Zanubrutinib Zanubrutinib dose to be reduced in case of concomitant
administration with moderate or strong CYP3A inhibitors
Not reported
Avoid coadministration with moderate or strong CYP3A inducers
Copanlisib Concomitant use with strong CYP3A inhibitors should be avoided.
Copanlisib dose to be reduced at 45 mg if a strong CYP3A
inhibitor must be used
Not reported
Strong CYP3A inducers should be avoided
Duvelisib Concomitant use with strong CYP3A inhibitors should be avoided.
Duvelisib dose to be reduced at 15 mg BID if a strong CYP3A
inhibitor must be used
May increase AUC of sensitive CYP3A4 substrates
Strong CYP3A inducers should be avoided
Venetoclax Strong CYP3A inhibitors should not be used during rampup.
Venetoclax dose to be reduced when strong inhibitors are used
at steady state
May increase warfarin Cmax and AUCinf resulting in increased risk
of bleeding. Venetoclax increases Cmax and AUCinf of Pgp
substrates
If moderate CYP3A or Pgp inhibitors are used during rampup or
at steady state, dose of venetoclax must be reduced
Concomitant use with strong or moderate CYP3A inducers should
be avoided
Abbreviations: CYP3A, cytochrome P450 3A; Pgp, P glycoprotein 1.
80
-
STATHIS
rampup dosing scheme. Therefore, the recommendation is to avoid
concomitant use with a strong CYP3A inhibitor at initiation and
during the rampup phase. During treatment at the steady dose,
alternative medications or dose adaptation of venetoclax and
frequent monitoring for adverse events should be considered. Strong
or moderate CP3A inducers can also result in changes and in
particular in decreased exposure to venetoclax and therefore current
recommendation is to avoid concomitant administration.
Finally, venetoclax may also alter the exposure to other drugs
and in particular it can increase warfarin levels and thus increase
the risk of bleeding. International normalized ratio should there-
fore be regularly checked in patients taking warfarin with
venetoclax.
6
|
CONCLUSION
Drug interactions can involve recently approved new drugs for pa-
tients with lymphoid malignancies. In particular, small molecules
targeted agents are primary metabolized by CAP3A and their coad-
ministration with strong or moderate CYP3A inhibitors or inducers
can result respectively in increased or decreased plasma concentra-
tions and therefore in risks of toxicity or decreased efficacy.
Awareness of this problem and a regular check of the medications of
the patients and consultation with pharmacists in case of any doubts
for potential DDIs could help to prevent these interactions and
especially those that could result in clinically significant conse-
quences for patients.
CONFLICT OF INTERESTS
Institutional grants for clinical trials: Merck, Bayer, Roche, Novartis,
Pfizer, ADC Therapeutics, MEW Pharma, Eli Lilly; advisory board:
Roche; consultant: Bayer, Eli Lilly; travel grant: PharmaMar, Abbvie.
DATA AVAILABILITY STATEMENT
Data sharing not applicable to this article as no datasets were
generated or analyzed during the current study.
REFERENCES
1. Byrd JC, Furman RR, Coutre SE, et al. Targeting BTK with ibrutinib
in relapsed chronic lymphocytic leukemia. N Engl J Med.
2013;369:3242.
2. Wang ML, Rule S, Martin P, et al. Targeting BTK with ibrutinib in
relapsed or refractory mantlecell lymphoma. N Engl J Med.
2013;369:507516.
3. Byrd JC, Harrington B, O'Brien S, et al. Acalabrutinib (ACP196) in
relapsed chronic lymphocytic leukemia. N Engl J Med. 2016;374:323
332.
4. Roberts AW, Davids MS, Pagel JM, et al. Targeting BCL2 with
venetoclax in relapsed chronic lymphocytic leukemia. N Engl J Med.
2016;374:311322.
5. Dreyling M, Morschhauser F, Bouabdallah K, et al. Phase II study of
copanlisib, a PI3K inhibitor, in relapsed or refractory, indolent or
aggressive lymphoma. Ann Oncol. 2017;28:21692178.
6. Sehn LH, Herrera AF, Flowers CR, et al. Polatuzumab vedotin in
relapsed or refractory diffuse large Bcell lymphoma. J Clin Oncol.
2020;38:155165.
7. Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR
Tcell therapy in refractory large Bcell lymphoma. N Engl J Med.
2017;377:25312544.
8. Schuster SJ, Bishop MR, Tam CS, et al. Tisagenlecleucel in adult
relapsed or refractory diffuse large Bcell lymphoma. N Engl J Med.
2019;380:4556.
9. Beijnen JH, Schellens JH. Drug interactions in oncology. Lancet
Oncol. 2004;5:489496.
10. König J, Müller F, Fromm MF. Transporters and drugdrug in-
teractions: important determinants of drug disposition and effects.
Pharmacol Rev. 2013;65:944966.
11. Scripture CD, Figg WD. Drug interactions in cancer therapy. Nat Rev
Cancer. 2006;6:546558.
12. Riechelmann RP, Del Giglio A. Drug interactions in oncology: how
common are they? Ann Oncol. 2009;20:19071912.
13. van Leeuwen RWF, Brundel DHS, Neef C, et al. Prevalence of po-
tential drugdrug interactions in cancer patients treated with oral
anticancer drugs. Br J Cancer. 2013;108:10711078.
14. Mouzon A, Kerger J, D'Hondt L, Spinewine A. Potential interactions
with anticancer agents: a crosssectional study. Chemotherapy.
2013;59:8592.
15. van Leeuwen RWF, Jansman FGA, van den Bemt PMLA, et al. Drug
drug interactions in patients treated for cancer: a prospective study
on clinical interventions. Ann Oncol. 2015;26:992997.
16. Marcath LA, Coe TD, Hoylman EK, et al. Prevalence of drugdrug
interactions in oncology patients enrolled on National Clinical Tri-
als Network oncology clinical trials. BMC Cancer. 2018;18:1155.
17. Bossaer JB, Eskens D, Gardner A. Sensitivity and specificity of drug
interaction databases to detect interactions with recently approved
oral antineoplastics. J Oncol Pharm Pract. 2021. https://doi.org/10.
1177/1078155220984244
18. Finnes HD, Chaffee KG, Call TG, et al. Pharmacovigilance during
ibrutinib therapy for chronic lymphocytic leukemia (CLL)/small
lymphocytic lymphoma (SLL) in routine clinical practice. Leukemia
Lymphoma. 2017;58:13761383.
19. Scheers E, Leclercq L, de Jong J, et al. Absorption, metabolism, and
excretion of oral 14C radiolabeled ibrutinib: an openlabel, phase
I, singledose study in healthy men. Drug Metab Dispos.
2015;43:289297.
20. de Jong J, Skee D, Murphy J, et al. Effect of CYP3A perpetrators on
ibrutinib exposure in healthy participants. Pharmacol Res Perspect.
2015;3:e00156.
21. de Zwart L, Snoeys J, De Jong J, Sukbuntherng J, Mannaert E,
Monshouwer M. Ibrutinib dosing strategies based on interaction
potential of CYP3A4 perpetrators using physiologically based
pharmacokinetic modeling. Clin Pharmacol Ther. 2016;100:548557.
22. IMBRUVICA (Ibrutinib)—FDA. https://www.accessdata.fda.gov/
drugsatfda_docs/label/2015/205552s002lbl.pdf
23. Kohrt HE, SagivBarfi I, Rafiq S, et al. Ibrutinib antagonizes
rituximabdependent NK cellmediated cytotoxicity. Blood.
2014;123:19571960.
24. Calquence—FDA. https://www.accessdata.fda.gov/drugsatfda_docs/
label/2017/210259s000lbl.pdf
25. Label—FDA. https://www.accessdata.fda.gov/drugsatfda_docs/label/
2019/213217s000lbl.pdf
26. Fancher KM, Pappacena JJ. Drug interactions with Bruton's tyrosine
kinase inhibitors: clinical implications and management. Cancer
Chemother Pharmacol. 2020;86:507515.
27. Stephens DM, Byrd JC. How I manage ibrutinib intolerance and
complications in patients with chronic lymphocytic leukemia. Blood.
2019;133:12981307.
STATHIS
-
81
28. Aliqopa—FDA. https://www.accessdata.fda.gov/drugsatfda_docs/
label/2017/209936s000lbl.pdf
29. Duvelisib—FDA. https://www.accessdata.fda.gov/drugsatfda_docs/
label/2018/211155s000lbl.pdf
30. Venclexta—FDA. https://www.accessdata.fda.gov/drugsatfda_docs/
label/2019/208573s013lbl.pdf
How to cite this article: Stathis A. New drugs and
pharmacological interactions in real life. Hematological Oncology.
2021;39(S1):7882. https://doi.org/10.1002/hon.2859
82
-
STATHIS
Received: 26 February 2021
DOI: 10.1002/hon.2855
SUPPLEMENT ARTICLE
Vulnerabilities in the tumor and microenvironment in
follicular lymphoma
Ferran AraujoAyala
1
|Patricia PérezGalán
1,2
|Elias Campo
1,2,3,4
1
Department of HematologyOncology,
Institut d'Investigacions Biomèdiques August
Pi I Sunyer (IDIBAPS), Barcelona, Spain
2
Hematological Neoplasms Program, Centro
de Investigación Biomédica en RedOncología
(CIBERONC), Madrid, Spain
3
Hematopathology Unit, Pathology
Department, Hospital Clínic of Barcelona,
Barcelona, Spain
4
Department of Basic Clinical Practice,
University of Barcelona, Barcelona, Spain
Correspondence
Elias Campo, Hospital Clinic of Barcelona,
Calle Villarroel 170, 08036Barcelona, Spain.
Email: ecampo@clinic.cat
Abstract
Follicular lymphoma (FL) is a paradigm of tumors that require the interaction be-
tween tumor and microenvironment cells to foster their development from initial
steps to progression. Recent largescale genome studies have uncovered multiple
genetic alterations of FL that influence the microenvironment in two main di-
rections, promoting tumor cell survival and proliferation and facilitating their
evasion from immune antitumor signals. Understanding the crosstalk between tu-
mor Bcells and the microenvironment will facilitate the identification of vulnera-
bilities that may offer novel targets for treatment of the patients. This review
highlights recent findings showing the effect of common genetic mutations modu-
lating the cell composition of the tumor microenvironment and the novel thera-
peutic perspectives to target these interactions.
KEYWORDS
follicular lymphoma, genomic alterations, microenvironment, target therapies
1
|
INTRODUCTION
Follicular lymphoma (FL) develops in the germinal center (GC) of
lymphoid follicles due, in most of the cases, to the acquisition of the t
(14;18) translocation in precursor cells of the bone marrow which
leads to the overexpression of the antiapoptotic protein BCL2. A
second early oncogenic event in FL is the introduction of somatic
mutations in the variable regions of immunoglobulin genes creating
novel Nglycosylation sites for highly mannosylated glycans that
directly interact with endogenous lectins found in cells of the tumor
microenvironment (TME). These interactions activate Bcell receptor
signals required for tumor development. The third pillar in early steps
of FL is recurrent mutations in epigenetic regulator genes that confer
selective growth advantages to the Bcell and promote favorable
interactions with the microenvironment.
1,2
Further, development and
progression of FL are associated with subsequent acquisition of
additional genomic alterations that target different pathways related
to cell differentiation, survival, proliferation, dissemination, and
metabolic advantages among others. In addition to the genomic and
epigenomic alterations, FL cells modulate the microenvironment to
promote the tumor cell growth.
3
Understanding these complex
phenomena and the crosstalk between tumor and stromal cells may
facilitate the identification of vulnerabilities that will offer novel
targets for treatment of the patients.
2
|
FL GENOME AND NEW VULNERABILITIES
2.1
|
Epigenetic modulators
Largescale genomic studies combined with functional analysis have
elucidated the mutational profile of FL and defined the several
altered pathways involved in the pathogenesis of these tumors. The
most common aberrations are mutations in the epigenetic regulators
KMT2D (60%–90%), CREBBP/EP300 (50%–70%/10%–20%), and
EZH2 (10%–30%). The high frequency of these lesions, their findings
in “in situ” follicular neoplasia and acquisition in later steps in the
evolution indicate that they are very early events but also favor the
progression of the tumors.
4,5
The loss of function mutations in
the histone H3K4 methyltransferase KMT2D and in the H3K27
Hematological Oncology. 2021;39(S1):8387. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
-
83
acetyltransferase CREBBP/EP300 together with the gain of function
in the H3K27 methyltransferase EZH2 tend to confer a repressive
functional state of the genes targeted by these chromatin modifiers
related to Bcell differentiation and cell cycle regulation that main-
tain tumor cells in a GC stage (Table 1).
1,5
Particularly, CREBBP mu-
tations silence genes that are direct targets of the BCL6HDAC3
oncorepressor complex, including those that regulate Bcell
signaling and immune responses, such as class II major histocom-
patibility complex (class II MHC). Other Bcell neoplasias, such as
Hodgkin's lymphoma and primary mediastinal large Bcell lymphoma
show class II MHC deregulation, but the mechanism is different
mainly associated with CIITA alterations.
6
CREBBPmutated tumors
also seem to have less helper and cytotoxic Tcells in the microen-
vironment suggesting that these mutations favor tumor cell immune
evasion.
5,6
These crucial alterations in FL pathogenesis support the
idea that CREBBP may be a gene with high therapeutic potential. In
this sense, HDAC3 inhibition restores in part the immune responses
and therefore may represent a new therapeutic approach in FL.
7
Recent studies have also linked EZH2 mutations to the reprograming
of the tumor and microenvironment interactions. Mutant cells in the
light zone of the GC seem to be less dependent on Tfollicular helper
(T
FH
) cells while potentiate their interaction and dependence on
follicular dendritic cells (FDCs).
8
On the other hand, EZH2 also seems
to play a role in the development of T and natural killer (NK) cells.
The potential benefit of EZH2 inhibitors independently of its
mutational status may be related to this extra tumor cell activity.
Interestingly, Food and Drug Administration has approved the use of
the selective EZH2 inhibitor tazemetostat for adult relapse/re-
fractory (R/R) patients with EZH2mutated tumors and patients with
R/R FL who have no satisfactory alternative treatment options.
2.2
|
Immune evasion
Mutations and deletions of TNFRSF14 in 1p36, also known as
herpesvirus entry mediator A are a common event in FL (50%).
1
TNFRSF14 is the ligand for BTLA expressed in TFH cells and induces
inhibitory signals on these cells. The disruption of these interactions
results in increased recruitment of the tumor supportive T
FH
and
release of cytokines that also favor a proTME. In addition,
TNFRSF14 generates inhibitory signals on the Bcell receptor (BCR)
of the Bcells that are released by the oncogenic inactivation. A
preclinical CART construct has been developed to continuously
produce soluble TNFRSF14 in the microenvironment that restores its
inhibitory function.
9
Ephrin receptor A7 (EPHA7) is a soluble tumor suppressor
inactivated in approximately 70% of FL by mutations and de-
letions. Loss of expression avoids its binding to EPHA2 receptor,
which inhibits extracellular regulated MAP kinase (ERK) and SRC
protooncogene, nonreceptor tyrosine kinase oncogenic signals.
TABLE 1Genomic and microenvironment alterations in follicular lymphoma as potential targets for novel treatment
Target Type of alteration Biological consequence Targeted drug Combinations Status
Genomic aberrations
CREBBP Lossoffunction mutations Downregulation of MHC class II BRD3308 AntiPDL1 Preclinical
EZH2 Gainoffunction mutations Increased dependency on FDC Tazemetostat None Approved
TNFRSF14 Inactivating mutations Increased T
FH
recruitment CART None Preclinical
RRAGC Activating mutations Insensitivity to nutrient
deprivation
Temsirolimus Bendamustine rituximab Clinical
Increased B cell responses
FLmicroenvironment crosstalk
Bcell receptor FLFDC BCR activation Ibrutinib Rituximab Clinical
FLDCSIGN
+
Mϕ
PI3KδFLT
FH
CD40CD40L activation, T
reg
recruitment
Idelalisib None Approved
FLT
reg
PI3Ky FLmonocyte/MϕMyeloid cell recruitment, M2
polarization
Duvelisib None Approved
CSF1R FLMϕMyeloid cell recruitment and
differentiation, M2
polarization
Pexidartinib Rituximab Preclinical
CD47SIRPαFLMϕCD47
+
FL cells inhibit
phagocytosis by Mϕand
neutrophils
AntiCD47 Rituximab Clinical
FLneutrophils
Abbreviations: BCR, Bcell receptor; FDC,follicular dendritic cell; FL, follicular lymphoma; MHC, major histocompatibility complex; T
reg
, Tregulatory
cell; T
FH
, Tfollicular helper.
84
-
ARAUJOAYALA ET AL.
Therapies that restore EPHA7 function or inhibit downstream
oncogenic signals activated due to EPHA7 loss might be useful to
treat FLmutated patients.
10
Recent studies have identified mutations in mammalian target of
rapamycin complex 1 (mTORC1) pathway. Particularly, RRAGC acti-
vating mutations are present in 17% of FL patients. These mutations
activate mTORC1 bypassing amino acid deprivation and also confer
an independent requirement of T
FH
cells for tumor survival.
Concordantly, these tumors have less T
FH
cells in their stroma.
11
Thus, the identification of patients carrying mutations in this meta-
bolic pathway may benefit of selective inhibitors.
Cathepsin S (CTSS) has been found to be mutated (6%) or
overexpressed (13%) in FL leading to its hyperactivation.
12,13
CTSS
cleaves CD74, among other targets, which is fundamental for MHC II
assembly and antigen (Ag) presentation. Hyperactive CTSS yields a
more efficient Agspecific CD4+T activation, increased CD4+Tcell
infiltration, and proinflammatory cytokine perturbation in FL mouse
models and human FL samples. Interestingly, the subversion of TME
by this alteration correlates with high PDL2 expression that yielded
FL patients more responsive to antiPD1 regimens, and also associ-
ates with better outcomes to immunochemotherapy (rituximab,
cyclophosphamide, doxorubicin, vincristine, and prednisone
[RCHOP]).
Other mutated genes in FL that also target the interactions be-
tween tumor cells and microenvironment are B2microglobulin and
CD58 that seem to be more common in transformed FL than in the
early steps of the disease.
1
Thus, reactivating their function may
improve cytotoxic response via CD8
+
T cells.
3
|
THERAPEUTIC OPPORTUNITIES FROM TME
Interaction of FL cells with nontumor cells constitutes a key feature
in the pathophysiology of the tumor. A precise characterization of the
TME could uncover new vulnerabilities to treat these patients. The
tumor niche of this lymphoma is composed of different cells types,
which create a tumor supportive environment and facilitate the scape
from the host antitumor immune responses. The main actors in this
TME are different subpopulations of Tcells, myeloid derived cells,
mainly tumorassociated macrophages (TAMs), and stroma cells,
mainly FDCs and cancerassociated fibroblasts, among others. These
cells communicate among them and with tumor cells through a
network of cytokines and celltocell interactions.
3.1
|
Tcells
Tcells are a heterogeneous group of cells that interplay with FL cells
at different levels mediating antitumor responses or, contrarily,
providing supportive protumoral signals. CD8
+
cytotoxic cells,
together with NK cells and probably Tγ/δcells, mediate antitumor
responses. However, FL cells may counteract this antitumor effect by
secreting interleukin 12 (IL12) that leads to an exhaustion of CD8
+
cells and by recruiting Tregulatory cells (Tregs) that inhibit
degranulation and cytotoxic activity of CD8
+
cells.
14
On the other
side of the balance, two subpopulations of CD4
+
cell, T
FH
, and Treg
play key roles in providing tumor support and facilitating immune
evasion. In this way, T
FH
stimulate tumor Bcells mediated by CD40L/
CD40 and MHC class II interactions.
15
These cells secrete IL4, which
triggers activation of Bcells mediated by ERK and STAT6, and the
chemokine CCL22 that recruits immunomodulatory Treg cells. Treg
cells are a subset of CD4
+
cells characterized by the expression of
the transcription factor FOXP3. These cells seem to play a protumor
role due to their immunosuppressive activity on CD4 and CD8 cells.
However, a subset of T follicular regulatory cells (Tfr) has been also
recognized by the additional expression of BCL6, CXCR5, ICOS, and
PD1. These cells limit the expansion of the GC reaction and down-
regulate the effects of T
FH
cells. However, there is still some con-
troversy regarding the impact of Tfr FOXP3+expression patterns on
FL survival.
16,17
The relevance of T
FH
and Treg cells in FL has been highlighted by
the effect of PI3Kδinhibitors disrupting the crosstalk between FL
cells and T
FH
cells. In vitro studies have shown that the PI3Kδin-
hibitor idelalisib, diminishes tumor cell proliferation, and reshapes
immune microenvironment inhibiting the recruitment of classical
Treg cells by downregulating CCL22.
18
Finally, in this context of B–Tcell crosstalk, FL cells present
immunoglobulin neoantigens that may play a determinant role in host
immune responses, and constitute potential immunotherapeutic
targets.
19
3.2
|
TAMs and other stromal cells
The role of TAMs in FL has been controversial. In the prerituximab
era, some studies suggested that macrophage infiltration correlated
with lower survival. Nevertheless, the addition of rituximab to che-
motherapeutical drugs modified their prognostic impact. The number
of infiltrating macrophages in patients treated with standard immu-
nochemoterapeutic regimens such as RCHOP was associated with
improved overall survival.
20
In addition to the influence of new therapies in the possible role
of TAM, different subpopulations of these cells may also play
different roles in the pathogenesis of FL. Macrophages polarized to a
M1 phenotype may exert antitumor properties by producing proin-
flammatory such as IL1, IL6, IL12, and tumor necrosis factor α,
whereas M2polarized macrophages are protumor since they are
able to downregulate MHC and IL12, as well as expressing anti
inflammatory molecules such as arginase and IL10, and the scav-
enger receptor CD163. Moreover, M2 macrophages are also involved
in tumor angiogenesis. The protumor effect of TAMs, particularly
with an M2 polarization, makes these cells and their interactions with
tumor cells an attractive target for therapies.
In is noteworthy, that contrary to other Bcell lymphomas such
as diffuse large Bcell lymphoma (DLBCL) or Hodgkin lymphoma, FL
cells do not express PDL1, and PDL2 is moderately expressed in a
ARAUJOAYALA ET AL.
-
85
high proportion of FL cases.
21
However, PD1 ligands are present in
the immune infiltrates where PDL1
+
histiocytes have been detected
in the Tcellrich zone of the neoplastic follicles,
22
justifying the
therapeutic targeting of this pathway.
A target to interfere the FLTAMs crosstalk is the colony
stimulating factor 1 receptor (CSF1R), also known as macrophage
colonystimulating factor receptor, as it is a relevant element in the
differentiation and survival of macrophages. Noteworthy, high levels
of CSF1R expression in FL have been associated with higher histo-
logical grade and risk of transformation suggesting that targeting
CSF1R may be relevant in highrisk patients. Although CSF1R is
expressed in both M1 and M2 subtypes, its inhibition by pexidartinib
preferentially diminish the viability of M2 macrophages and repo-
larize them to M1 macrophages suggesting that it may reeducated
TAMs toward an antitumor phenotype.
CD47 is a receptor usually expressed in cancer cells that pre-
vents phagocytosis forming a complex with signalregulatory protein
α(SIRPα). Among other phagocytes (e.g., neutrophils), macrophages
may express SIRP‐α in their membrane compromising their antitumor
phagocytic function when interacting with CD47positive lymphoma
cells (Table 1). By administrating a therapeutic antibody that blocks
CD47, phagocytosis of tumor cells is increased and adaptive immu-
nity is enhanced.
23
This effect has been explained by the Ag
presenting function of macrophages. After phagocytizing tumor
cells, macrophages may present tumor Ag to CD4
+
Thelper cells
triggering an antitumor response. Altogether, antiCD47 antibodies
are now in clinical trials as promising drug candidates to activate the
immune system in FL, especially in combined immunotherapeutic
regimens including rituximab.
PI3Kγis expressed by microenvironment cells that support tu-
mor growth such as CD4 and M2 macrophages. The dual PI3Kγδ
inhibitor duvelisib exerts antitumor effects by targeting both the
tumor and microenvironment cells (Table 1). This compound inhibits
tumor cell proliferation and survival whereas also promotes the dif-
ferentiation of M2like TAMs to a M1 phenotype and inhibits the
antitumor effect of Tcells interfering with the tumor supporting
properties of the TME. Other drugs that interfere with BCR pathway
such as ibrutinib may also affect the crosstalk of tumor cells and
macrophages. While BTK inhibitors (BTKi) have not showed optimal
results as monotherapy, the combination of the BTKi ibrutinib with
the antiCD20 rituximab has improved clinical trials results.
24
FDCs are from mesenchymal origin and build a network to
support the GC reaction. They are able to present antibody—Ag
complexes on their cell surface engaging BCR activation, survival
of malignant cells and recruitment of T
FH
cells.
18
Moreover, we
have recently demonstrated that FLFDC crosstalk induces mono-
cyte recruitment and their differentiation to M2protumoral
macrophages.
25
Other cell types that may play an important role in the disease
are different subsets of Tcells, TAMs, mesenchymal stem cells, or
follicular reticular cells, among others. Interestingly, the evolution of
this TME might be determinant in the transformation of FL to DLBCL.
4
|
CONCLUDING REMARKS
Although in most cases FL initially presents as an indolent disease, a
significant proportion of cases are primary refractory to standard
treatment (RCHOP or derivatives) and a high percentage of those
who initially respond to treatment will eventually relapse. Once the
relapse occurs, the prognosis of the patients worsens, especially in
those who suffer an early relapse, within the first 24 months of
treatment progression of disease within 2 years. Furthermore, those
patients are in higher risk of histological transformation to an
aggressive lymphoma, mainly DLBCL. The evolution of the patients
cannot be precisely predicted exclusively based on single genetic
alterations. Recent studies have shown that assays that include
multiple genetic aberrations or combining both tumor Bcell biology
and tumor microenvironment alterations may be better predictive
models. These findings reinforce the need to integrate the mutational
profile of the tumor cells together with the complex interactions of
the TME to predict the biological risk of the patients. This compre-
hensive perspective should assist in identifying tumor and microen-
vironment vulnerabilities that will allow treating them effectively and
avoiding early clinical progression.
ACKNOWLEDGMENTS
This study was funded by Ministerio de Ciencia e Innovación, Grant
No. RTI2018094274BI00 (to E. C.) and Fundacio la Marató TV3
(TAIFOL 20193330) (to P. PG.).
CONFLICT OF INTERESTS
The authors declare that there are no conflict of interests.
DATA AVAILABILITY STATEMENT
Data sharing not applicable to this article as no datasets were
generated or analyzed during the current study.
REFERENCES
1. Huet S, Sujobert P, Salles G. From genetics to the clinic: a trans-
lational perspective on follicular lymphoma. Nat Rev Cancer.
2018;18:224239.
2. Küppers R, Stevenson FK. Critical influences on the pathogenesis of
follicular lymphoma. Blood. 2018;131:22972306. https://doi.org/10.
1182/blood201711764365
3. Kridel R, Sehn LH, Gascoyne RD. Pathogenesis of follicular lym-
phoma. J Clin Invest. 2012;122:34243431. https://doi.org/10.1172/
JCI63186
4. Schmidt J, RamisZaldivar JE, Bonzheim I, et al. CREBBP gene mu-
tations are frequently detected in in situ follicular neoplasia. Blood.
2018;132:26872690.
5. Green MR. Chromatin modifying gene mutations in follicular lym-
phoma. Blood. 2018;131:595604. https://doi.org/10.1182/blood
201708737361
6. Green MR, Kihira S, Liu CL, et al. Mutations in early follicular lym-
phoma progenitors are associated with suppressed antigen presen-
tation. Proc Natl Acad Sci USA. 2015;112:E1116E1125. https://doi.
org/10.1073/pnas.1501199112
7. Mondello P, Tadros S, Teater M, et al. Selective inhibition of HDAC3
targets synthetic vulnerabilities and activates immune surveillance
86
-
ARAUJOAYALA ET AL.
in lymphoma. Cancer Discov. 2020;10:440459. https://doi.org/10.
1158/21598290.CD190116
8. Béguelin W, Teater M, Meydan C, et al. Mutant EZH2 induces a pre
malignant lymphoma niche by reprogramming the immune response.
Cancer Cell. 2020;37:655673.e11. https://doi.org/10.1016/j.ccell.
2020.04.004
9. Boice M, Salloum D, Mourcin F, et al. Loss of the HVEM tumor
suppressor in lymphoma and restoration by modified CART cells.
Cell. 2016;167:405418.e13. https://doi.org/10.1016/j.cell.2016.
08.032
10. Oricchio E, Nanjangud G, Wolfe AL, et al. The Ephreceptor A7
is a soluble tumor suppressor for follicular lymphoma. Cell.
2011;147:554564. https://doi.org/10.1016/j.cell.2011.09.035
11. OrtegaMolina A, DeleytoSeldas N, Carreras J, et al. Oncogenic
Rag GTPase signalling enhances B cell activation and drives follic-
ular lymphoma sensitive to pharmacological inhibition of mTOR.
Nat Metab. 2019;1:775789. https://doi.org/10.1038/s42255019
00988
12. Dheilly E, Battistello E, Katanayeva N, et al. Cathepsin S regulates
antigen processing and T cell activity in nonHodgkin lymphoma.
Cancer Cell. 2020;37:674689.e12. https://doi.org/10.1016/j.ccell.
2020.03.016
13. Bararia D, Hildebrand JA, Stolz S, et al. Cathepsin S alterations
induce a tumorpromoting immune microenvironment in follicular
lymphoma. Cell Rep. 2020;31:107522. https://doi.org/10.1016/j.
celrep.2020.107522
14. AméThomas P, Tarte K. The yin and the yang of follicular lymphoma
cell niches: role of microenvironment heterogeneity and plasticity.
Semin Cancer Biol. 2014;24:2332. https://doi.org/10.1016/j.sem
cancer.2013.08.001
15. AméThomas P, Le Priol J, Yssel H, et al. Characterization of intra-
tumoral follicular helper T cells in follicular lymphoma: role in the
survival of malignant B cells. Leukemia. 2012;26:10531063. https://
doi.org/10.1038/leu.2011.301
16. Carreras J, Lopezguillermo A, Fox BC, et al. High numbers of
tumorinfiltrating FOXP3positive regulatory T cells are associated
with improved overall survival in follicular lymphoma. Blood.
2006;108:29572964. https://doi.org/10.1182/blood200604018
218.E.C
17. Farinha P, AlTourah A, Gill K, Klasa R, Connors JM, Gascoyne RD.
The architectural pattern of FOXP3positive T cells in follicular
lymphoma is an independent predictor of survival and histologic
transformation. Blood. 2010;115:289295. https://doi.org/10.1182/
blood200907235598
18. Serrat N, GuerreroHernández M, Hernández H, et al. PI3Kd inhi-
bition reshapes follicular lymphomaimmune microenvironment
cross talk and unleashes the activity of venetoclax. Blood Adv. 2020;
4(17):4217–4231. https://doi.org/10.1182/bloodadvances.202000
1584
19. Khodadoust MS, Olsson N, Chen B, et al. Bcell lymphomas present
immunoglobulin neoantigens. Blood. 2019;133:878881.
20. Kridel R, Xerri L, GelasDore B, et al. The prognostic impact of
CD163positive macrophages in follicular lymphoma: a study from
the BC Cancer Agency and the Lymphoma Study Association. Clin
Cancer Res. 2015;21:34283435. https://doi.org/10.1158/1078
0432.CCR143253
21. Laurent C, Charmpi K, Gravelle P, et al. Several immune escape
patterns in nonHodgkin's lymphomas. Oncoimmunology. 2015;4:
e1026530. https://doi.org/10.1080/2162402X.2015.1026530
22. Myklebust JH, Irish JM, Brody J, et al. High PD1 expression and
suppressed cytokine signaling distinguish T cells infiltrating follicular
lymphoma tumors from peripheral T cells. Blood. 2013;121:1367
1376. https://doi.org/10.1182/blood201204421826
23. Zhang W, Huang Q, Xiao W, et al. Advances in antitumor treat-
ments targeting the CD47/SIRPαaxis. Front Immunol. 2020;11:18.
https://doi.org/10.3389/fimmu.2020.00018
24. Fowler NH, Nastoupil L, De Vos S, et al. The combination of ibrutinib
and rituximab demonstrates activity in firstline follicular lymphoma.
Br J Haematol. 2020;189:650660. https://doi.org/10.1111/bjh.
16424
25. Valero JG, MatasCespedes A, Rodriguez V, et al. The receptor of
the colonystimulating factor1 (CSF1R) is a novel prognostic factor
and therapeutic target in follicular lymphoma. Leukemia. 2021.
https://doi.org/10.1038/s41375021012019
How to cite this article: AraujoAyala F, PérezGalán P,
Campo E. Vulnerabilities in the tumor and microenvironment
in follicular lymphoma. Hematological Oncology.
2021;39(S1):8387. https://doi.org/10.1002/hon.2855
ARAUJOAYALA ET AL.
-
87
Received: 26 February 2021
DOI: 10.1002/hon.2852
SUPPLEMENT ARTICLE
Upfront identification of highrisk follicular lymphoma
Carla Casulo
Department of Medicine, Division of
Hematology and Oncology, Wilmot Cancer
Institute, Rochester, New York, USA
Correspondence
Carla Casulo, Department of Medicine,
Division of Hematology and Oncology, Wilmot
Cancer Institute, 601 Elmwood Avenue Box
704, Rochester, NY 14642, USA.
Email: Carla_casulo@urmc.rochester.edu
Abstract
Follicular lymphoma (FL) is a common disease with clinically indolent behavior, and
a long natural history for the majority of patients. Despite excellent therapeutic
strategies currently available for FL, approximately 10%–20% of patients will
experience early disease progression, defined as occurring within two years of
diagnosis. These patients have poor outcomes, with overall survival at 5 years
ranging between 37% and 50%. Much of the biology driving early progression and
inferior survival is attributed to early transformation events; however, trans-
formation alone does not account for all the observed clinical heterogeneity and
survival differences among patients. Several clinical, genetic, and molecular alter-
ations in FL have been discovered that help define subsets of patients at risk for
multiply relapses and refractory disease, and are slowly making their way into risk
calculators to be used in daily practice. Additionally, the role of functional imaging
with PET scan, as well as circulating and cell free tumor DNA are being evaluated as
tools to define highrisk subsets of patients with FL. This review seeks to provide an
over view of current and evolving biomarkers that define highrisk FL at diagnosis.
The goal is for these tools to assist clinicians in integrating these rapidly evolving
prognosis models into clinical practice, in the hopes of riskstratifying treatments
and improving outcomes for patients.
KEYWORDS
follicular lymphoma, highrisk disease, relapse
1
|
INTRODUCTION
Follicular lymphoma (FL) arises from precursor B cells that acquire
the t(14;18), undergo additional genetic alterations, and develop
malignant transformation within the germinal center of the lymph
node.
1
Affecting approximately 2–4 people per 100,000 person
years, FL is the most common indolent lymphoma diagnosed in the
United States and Europe.
2
On a global scale, the incidence of FL
appears to be rising in developed countries, suggesting possible
environmental contributions affecting pathobiology and genetic
susceptibility. FL is described as clinically and biologically heteroge-
nous, owing to rare morphologic FL subtypes including pediatric type
and duodenal type FL variants, and distinct clinical behavior leading
to divergent outcomes in 20% of cases.
3
FL clinical behavior is generally indolent, where 80% of patients
experience longterm disease control following standard chemo
immunotherapy combinations. However, 20% or so of cases experi-
encing early disease progression have poor outcomes (5 years overall
survival [OS] of 35%–50%).
4
An additional subset of patients are
refractory to both antiCD20 antibodies and alkylators (so called
“double refractory”), have continuous relapses over time, or undergo
transformation to aggressive histology. These patients do not enjoy
the long natural history of FL with median OS of nearly 2 decades
and are often designated as “high risk,” owing to a more aggressive
clinical phenotype and less favorable outcomes.
5
Identifying patients with highrisk FL upfront is a global research
effort aimed at minimizing risk of death, histologic transformation,
and optimizing duration of treatment response to ease suffering and
88
-
Hematological Oncology. 2021;39(S1):8893. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
morbidity from this disease. The subject of this review is to discuss
current biomarkers and those in development that can aide in iden-
tifying highrisk FL.
1.1
|
Current biomarkers identifying highrisk
patients
Over the last several years, the discovery of novel clinical and
biologic markers has revolutionized FL disease prognostication and
risk precision.
6
However, despite wellestablished clinical calcula-
tors and molecular classifiers, the precise utilization of these tools
remains in question as it relates to selection of therapy at diagnosis
or relapse. When discussing biomarkers and their utilization to
identify subsets of patients with highrisk disease, it is relevant to
define the differences between prognostic and predictive markers.
7
Prognostic biomarkers appraise risk of a particular outcome such as
disease progression or death. Prognostic markers can assist in
developing riskadapted treatments. In contrast, predictive bio-
markers are reserved to associate a specific therapy with a
particular clinical response, or lack thereof. As such, identifying
disease response to specific therapies is enhanced with predictive
biomarkers. Complicating matters is that in many cases, some bio-
markers are both prognostic and predictive. In FL, the most well
established prognostic markers of outcome include clinical calcula-
tors such as the FLIPI or FLIPI2, histologic variations such as grade,
diseasespecific markers such as tumor burden and stage, and sur-
rogates for tumor kinetics such as early progression of disease
(POD24). Molecular markers such as the presence of somatic mu-
tations, and differential expression of genes within and outside the
FL tumor microenvironment, are gaining widespread attention as
critical determinants of disease outcome. Many studies have
established that functional imaging with PET at the end of therapy
also predicts both progression free (PFS) and overall survival (OS)
irrespective of treatment.
1.2
|
Clinical risk calculators
The FL International Prognostic Index (FLIPI) is the most commonly
used clinical calculator to determine FL prognosis.
8
The FLIPI's five
pretreatment patient characteristics (nodal sites, lactate dehydro-
genase, age >60 years, Stage IIIIV, and hemoglobin <12 g/dl) were
strongly prognostic of OS in over 900 patients with lowgrade FL
treated with chemotherapy (no immunotherapy) strongly associated
with 5and 10year OS based on low, intermediate risk, or highrisk
designation (Table 1). In a retrospective analysis of rituximab treated
patients by the German Low Grade Lymphoma Study Group (GLSG),
patients with highrisk FLIPI had inferior time to treatment failure
(67% at 2 years) compared to those with intermediate risk and low
risk FLIPI groups (90% and 92%, respectively).
9
Similarly, The Na-
tional LymphoCare Study, a prospective, multicenter observational
cohort of patients with FL in the United States including nearly 70%
of rituximab treated patients, observed a hazard ratio for OS higher
in the highrisk FLIPI group compared to the intermediate risk FLIPI
groups.
10
The FLIPI 2 calculator was another method used to predict PFS
upfront using additional parameters such as increased beta 2
microglobulin (B2M) and bone marrow involvement, and had com-
parable predictive ability as the FLIPI.
11
When attempting to spe-
cifically identify patients with POD24, data using the FLIPI and FLIP2
are more limited; however, a study by Jurinovic et al.
12
reported a
sensitivity between 70% and 78%, and a specificity of 56%–58%,
suggesting a heterogeneity among the POD24 patients that is inad-
equately captured by this prognostic index.
Other clinical indices have attempted to capture highrisk pa-
tients, by using a parsimonious mode incorporating only bone
marrow involvement and B2M level.
13
The PRIMAPI initially
included patients from the randomized PRIMA study of patients
treated with chemoimmunotherapy, with or without rituximab
maintenance. Highrisk patients had elevated B2M greater than
3 mg/L and had 5year PFS of only 37%. Other groups validated the
utility of the PRIMA PI in chemotherapyfree approaches using
immunomodulator lenalidomide on the RELVANCE study, and using
rituximab and interferon in the Nordic Lymphoma Group studies. The
PRIMAPI was not able to sufficiently discriminate patients treated
with lenalidomide, but did identify a highrisk group of patients with
FL experiencing shortened OS and time to treatment failure. While
authors did not explicitly evaluate the PRIMA to assess POD24,
there was no difference in transformation rates between both
groups, and they did find improved risk stratification compared to the
FLIPI, especially in patients older than 60 years.
14
The FLEX calculator was developed from patients treated on the
phase 3 GALLIUM trial (NCT01332968) using Obinutuzumab based
chemoimmunotherapy combinations in previously untreated
advanced FL. This novel prognostic model sough to identify highrisk
patients compared to FLIPI, FLIPI2, and PRIMAPI. The primary
endpoint was PFS but OS and risk of POD24 were evaluated. It
included nine clinical factors (male sex, sum of the product diameter
[SPD] in the highest quartile, Grade 3A histology, >2 extranodal sites,
ECOG PS >1, hemoglobin <12 g/dl, elevated B2M, elevated LDH,
and peripheral blood absolute natural killer cell count <100/μl).
15
The model was validated in the SABRINA trial (randomized
Phase 3 trial of subcutaneous rituximab vs. intravenous rituximab for
firstline FL treatment; NCT01200758). Lowrisk patients had zero to
two factors; and highrisk patients had three to nine factors. PFS at 3
years was 86% for highrisk patients and 68% for lowrisk patients.
Using FLEX, the sensitivity for a highrisk score to predict POD24
was 60%, versus 53% for FLIPI and FLIPI2, and 69% for PRIMAPI.
Specificity for POD24 was 68% with FLEX compared with 59% for
FLIPI and FLIPI2, and 47% for PRIMAPI (Table 1).
Compared to other frequently used clinical prognostic calcula-
tors, the FLEX appeared to have greater accuracy for predicting
POD24, particularly with novel immunotherapy combinations, and
the first prognostic score in FL developed in patients treated with
bendamustineand obinutuzumabbased regimens.
CASULO
-
89
1.3
|
Impact of the tumor microenvironment on
identifying highrisk FL
The microenvironment surrounding FL tumor cells (tumor micro-
environment [TME]) is a critically influential component of FL
tumorigenesis and clinical behavior. In 2004, the Leukemia and
Lymphoma Molecular Profiling Project described pivotal findings
using gene expression profiling on biopsy samples from patients
treated in the prerituximab era.
16
Two combinations of gene
expression were assessed for ability to predict survival, where
increased expression of genes in Tcells and macrophages had a
more favorable prognosis (relative risk of death 0.15). In com-
parison, increased expression of genes from macrophages or
dendritic cells had higher risk disease, with an unfavorable prog-
nosis (relative risk of death 9.35). Subsequent studies have been
unable to consistently reproduce this same impact of the immune
system's effect of FL prognosis, largely due to heterogeneously
treated patient populations and study small sizes. Kridel et al.
17
reported on the adverse prognostic impact of CD163 +macro-
phages, only in patients treated with rituximab and cyclophos-
phamide, vincristine, prednisone (RCVP), but not rituximab with
cyclophosphamide, doxorubicin, vincristine, and prednisone (R
CHOP). Similarly, other studies assessing the impact of number,
distribution, and frequency of tumorassociated macrophages and
tumorassociated T cells have resulted in inconsistent conclusions
on FL prognosis.
TABLE 1Prognostic calculators in follicular lymphoma
Risk model Factors Risk groups Prognostic impact
Sensitivity of
highrisk group
for predicting
POD24
Specificity of
highrisk group
for predicting
POD24
FLIPI Age >60
Stage IIIIV
Hg <12 g/dl
Elevated LDH
>Four nodal sites
Low risk(0–1 points) 91% 5year OS 53%–78% 56%–62%
Intermediate risk(2 points) 78% 5year OS
High risk(3 or more points) 53% 5year OS
FLIPI2 Age >60
Bone marrow
involvement
Hg <12 g/dl
Elevated β2
microglobulin
Mass >6 cm
Low risk(0–1 points) 80% 5year OS 53% 59%–76%
Intermediate risk(2 points) 51% 5year OS
High risk(3 or more points) 19% 5year OS
PRIMAPI β2 microglobulin >3g/L
Bone marrow
involvement
Low risk(0 points) 69% 5year PFS 69% 48%
Intermediate risk(1 point) 55% 5year PFS
High risk(2 points) 37% 5year PFS
FLEX Male sex
SPD in the highest
quartile
Histologic grade 3A
>2 extranodal sites
ECOG PS >1
Hg <12 g/dl
Elevated β2
microglobulin
NK cell count >100/μl
Elevated LDH
Low risk(0–2 points) 86% 3year PFS 60% 68%
High risk(3–9 points) 68% 3year PFS
M7FLIPI ECOG PS >1
FLIPI high risk
Mutations in EP300,
CREBBP, CARD11,
MEF2B, EZH2,
ARID1A, FOXO1
Low risk 68%–77% 5year FFS 43%–61% 77%–86%
High risk 22%–38% 5year FFS
POD24PI ECOG PS >1
FLIPI high risk
Mutations in EP300,
EZH2, FOXO1
Low risk 72%–77% 5year FFS 54%–78% 67%–73%
High risk 36%–50% 5year FFS
Tumor
microenvironment
PDL2 expression Immune infiltration
LO
Not reported Not reported
Immune infiltration
HI
90
-
CASULO
To identify the role of the TME in predicting the risk of POD24,
Tobin et al.
18
recently targeted gene sequencing using NanoString
technology from paraffin embedded tissue and multispectral immu-
nofluorescence on a tissue microarray, and applied them to two
groups: a discovery cohort of 132 patients from Princess Alexandria
Hospital with early and advanced staged follicular lymphoma who
received either chemotherapy or observation, and two independent
validation cohorts of 198 patients with advanced stage disease
treated with RCHOP and RCVP from the German lowgrade lym-
phoma study group and the British Columbia Cancer Agency. They
also performed T cell repertoire analysis, flow cytometry, immuno-
fluorescence, and next generation sequencing. Gene expression
profiling revealed distinct clustering of follicular lymphoma samples
based on high or low expression of immune infiltrating cells. Low
expression of certain immune markers (including PDL2, TNF‐α, CD4,
and CD68) were all associated with poor outcome; however, the
most specific marker with the highest specificity and sensitivity was
PDL2. They then dichotomized PDL2 expression into immune
infiltration “high” and “immune infiltration low” in subsequent ana-
lyses. PDL2 is an immune checkpoint present broadly on both tumor
cells and the tumor microenvironment. To localize its distribution,
they performed flow cytometry in fresh FL samples, and quantified
PDL2 expression by PCR. They identified that PDL2 gene expression
was distributed in both CD20+tumor cells as well as non CD20+
cells in the tumor microenvironment. However, the proportion of
PDL2 was lower in the CD20+cells. Overall there was lower
expression of all immune cells in the immune infiltration low
phenotype compare to the immune infiltration high phenotype. When
testing the relevance of immune infiltration and POD24, they found
that POD24 events in the Princess Alexandria discovery set were
more enriched for the immune infiltration low phenotype. These
findings were also validated in the British Columbia cancer agency
and German lymphoma study group populations. Nearly 50% of pa-
tients with low PDL2 had POD24 events compared to 16% in those
with high PDL2, concluding that low PDL2 identifies a subset of
patients enriched for POD24.
18
1.4
|
Somatic mutations, gene expression profiling,
and clinicpathologic models
Toward a precision approach, investigators from the GLSG harmo-
nized clinical and pathologic data to create a clinicogenetic risk
model aimed at more accurate risk prognostication in patients
receiving front line chemoimmunotherapy.
19
They performed deep
DNA sequencing from formalin fixed pretreatment biopsies to
analyze the mutational status of genes in 151 patients with follicular
lymphoma tumor samples. The resulting prognostic tool, called the
m7FLIPI, distilled down 74 genes into seven genes with nonsilent
mutations occurring at a variant allele frequency of 10% or greater,
and combined these with highrisk FLIPI status and ECOG perfor-
mance status. These included genes that increased risk of progres-
sion, including EP300, FOX01, CREBBP, CARD11, and those that
decreased risk of progression, including EZH2, ARID1A, and MEF2B.
The cumulative risk score was calculated by combining relative
weights of these genes in a multivariate analysis predicting failure
free survival. This m7FLIPI score was tested to identify POD24
but only captured about 50% of patients as high risk. Jurinovic et al.
12
later developed the POD24PI, designed to augment the sensitivity
and specificity of the m7FLIPI in early progressing patients. This
model included only three genes, including EP300, FOX01, and EZH2,
performance status, and FLIPI score. The POD24PI was more sen-
sitive at identifying POD24 patients but did not outperform other
metrics due to lower specificity.
12
A recently developed prognostic model, termed the BioFLIPI,
integrated intrafollicular CD4 expression with FLIPI to improves
identification of FL patients at risk for early treatment failure.
Mondello et al.
20
evaluated 496 patients using tissue microarrays for
CD4, CD8, FOXP3, CD32b, CD14, CD68, CD70, SIRPa, TIM3, PD1,
and PDL1. They found CD4, CD8, FOXP3, PD1, and SIRPαwere
associated with risk of early failure and were subsequently evaluated
in the validation set; however, only intrafollicular CD4 expression
remained significant in multivariate analysis. Lack of intrafollicular
CD4 expression was associated with higher risk of early failure in
both the discovery and validation cohorts, with a pooled OR =2.29
(95% CI 1.47–3.58; p<0.001). The BioFLIPI then combined CD4
intrafollicular expression and FLIPI into a onetofour scale and was
robustly associated with risk of early failure.
20
Gene expression signatures beyond those identified in the TME
have clinical utility on the upfront identification of patients with poor
outcomes in FL. While not yet ready for standard use, these molec-
ular signatures are highly promising for further investigation. Huet
et al.
21
performed a largescale gene expression profiling study using
samples from 160 untreated FL patients from the PRIMA study in FL
to establish define correlates of PFS and early treatment failure. The
authors identified 23 genes associated with the TME and Bcell
biology. These were strongly associated with disease progression
based on a score of low risk or high risk. Further adjusting for clinical
prognostic tools such as the FLIPI, use of maintenance rituximab, the
gene signature sustained its predictive utility.
21
This method pre-
dicted POD24 with a sensitivity of 43% and a specificity of 79%.
1.5
|
PETbased functional imaging biomarkers
Metabolic assessment of FL disease burden is considered standard of
care following firstline chemoimmunotherapy. It is well established
that attainment of complete metabolic response on a fivepoint scale
(ref) is highly prognostic of outcome, regardless of treatment used. A
pooled analysis of European patients treated on three prospective
clinical trials established that a positive PET scan (defined as Deau-
ville score of 4 or 5) following frontline treatment was highly
prognostic of both PFS and OS.
22
A novel method of metabolic ac-
tivity has gaining attention is total metabolic tumor volume (TMTV).
Defined as a cut off of 510 cm
3
or greater, high TMTV at diagnosis is
associated with a poor prognosis.
23
When combined with circulating
CASULO
-
91
tumor cells and cell free DNA, TMTV at diagnosis has also been
strongly correlated with inferior PFS.
24
2
|
WHAT ARE HOPES FOR THE FUTURE TO
IDENTIFY HIGHRISK FL UP FRONT?
2.1
|
Implementation of biomarkers into clinical
practice
There are currently a multitude of clinical, molecular, clinic
pathologic, and functional imaging based biomarkers that can group
patients with FL into several risk categories at the time of diagnosis.
These tools do not yet inform on how best to treat patients, which
remains a significant limitation to daytoday practical implementa-
tion. Additional research is needed to integrate these important de-
velopments in clinically meaningful ways. A promising strategy may
involve use minimal residual disease (MRD) testing with circulating
tumoror cellfree DNA. This has been used by DelfaueLarue and
colleagues in combination with TMTV to identify groups of patients
with FL at risk of poor outcomes (PFS of 65% at 4 years). Several
other studies demonstrate value in MRD assessment after chemo-
immunotherapy as high prognostic of end of treatment outcome, and
others also determine high levels of circulating tumor DNA as asso-
ciated with poor PFS.
24
Use of predictive markers will be especially relevant as the
design of precision, risk adapted therapies are developed. Gain of
function mutations in enhancer of zeste homolog 2 (EZH2) are
associated with favorable outcome in FL and occur in approximately
20% of individuals. The first in class oral inhibitor of EZH2 tazeme-
tostat was recently approved by the United States Food and Drug
Administration for patients with FL having an EZH2 mutation or
those with no other treatment options available. EZH2 mutation in Fl
diagnostic samples has been shown to be a favorable prognostic
marker; and also appears to be correlate with response to therapy. In
the pivotal Phase 2 study, patients harboring mutation in EZH2 had
higher objective response rates compared to patients with wild type
EZH2 (69% vs. 35%).
25
Moreover, a recent publication by Jurinovic
et al.
26
suggested that patients with an EZH2 mutation had a longer
PFS if they were treated with CHOP or CVP in the frontline setting
versus bendamustinebased therapy.
We look forward to the opportunity to risk adapt therapy by using
precision approaches for patient with FL. These and other studies
support the optimism that the ability to do so is on the horizon.
CONFLICT OF INTEREST
Carla Casulo received research funding from Gilead, BMS, and
Genentech.
REFERENCES
1. Carbone A, Roulland S, Gloghini A, et al. Follicular lymphoma. Nat
Rev Dis Primers. 2019;5(1):83. https://doi.org/10.1038/s41572019
0132x.
2. Teras LR, DeSantis CE, Cerhan JR, Morton LM, Jemal A, Flowers CR.
US lymphoid malignancy statistics by World Health Organization
subtypes. CA Cancer J Clin. 2016;66:443459. https://doi.org/10.
3322/caac.21357.
3. Swerdlow SH, Campo E, Pileri SA, et al. The 2016 revision of the
World Health Organization classification of lymphoid neoplasms.
Blood. 2016;127(20):23752390. https://doi.org/10.1182/blood
201601643569.
4. Casulo C, Day B, Dawson KL, et al. Disease characteristics, treat-
ment patterns, and outcomes of follicular lymphoma in patients 40
years of age and younger: an analysis from the National Lymphocare
Study. Ann Oncol. 2015;26(11):23112317. https://doi.org/10.1093/
annonc/mdv375.
5. Casulo C, Nastoupil L, Fowler NH, Friedberg JW, Flowers CR. Unmet
needs in the firstline treatment of follicular lymphoma. Ann Oncol.
2017;28(9):20942106. https://doi.org/10.1093/annonc/mdx189.
6. Welaya K, Casulo C. Follicular lymphoma: redefining prognosis,
current treatment options, and unmet needs. Hematology/Oncology
Clinics of North America. 2019:33;627638. https://doi.org/10.1016/j.
hoc.2019.03.003.4
7. Henry NL, Hayes DF. Cancer biomarkers. Mol Oncol. 2012;6(2):
140146. https://doi.org/10.1016/j.molonc.2012.01.010.
8. SolalCeligny P, Roy P, Colombat P, et al. Follicular lymphoma in-
ternational prognostic index. Blood. 2004;104(5):12581265. https://
doi.org/10.1182/blood2003124434.
9. Buske C, Hoster E, Dreyling M, Hasford J, Unterhalt M,
Hiddemann W. The Follicular Lymphoma International Prognostic
Index (FLIPI) separates highrisk from intermediateor lowrisk
patients with advancedstage follicular lymphoma treated front
line with rituximab and the combination of cyclophosphamide,
doxorubicin, vincristine, and prednisone (RCHOP) with respect to
treatment outcome. Blood. 2006;108(5):15041508. https://doi.org/
10.1182/blood200601013367.
10. Nooka AK, Nabhan C, Zhou X, et al. Examination of the follicular
lymphoma international prognostic index (FLIPI) in the National
LymphoCare study (NLCS): a prospective US patient cohort treated
predominantly in community practices. Ann Oncol. 2013;24(2):
441448. https://doi.org/10.1093/annonc/mds429.
11. Federico M, Bellei M, Marcheselli L, et al. Follicular lymphoma in-
ternational prognostic index 2: a new prognostic index for follicular
lymphoma developed by the international follicular lymphoma
prognostic factor project. J Clin Oncol. 2009;27(27):45554562.
https://doi.org/10.1200/JCO.2008.21.3991.
12. Jurinovic V, Kridel R, Staiger AM, et al. Clinicogenetic risk models
predict early progression of follicular lymphoma after firstline
immunochemotherapy. Blood. 2016;128(8):11121120. https://doi.
org/10.1182/blood201605717355.
13. Bachy E, Maurer MJ, Habermann TM, et al. A simplified scoring
system in de novo follicular lymphoma treated initially with immu-
nochemotherapy. Blood. 2018;132:49. https://doi.org/10.1182/
blood201711816405.
14. Kimby E, Lockmer S, Holte H, et al. The simplified follicular lym-
phoma PRIMAprognostic index is useful in patients with firstline
chemofree rituximabbased therapy. Br J Haematol. 2020;191:738.
https://doi.org/10.1111/bjh.16692.
15. Mir F, Mattiello F, Grigg A, et al. Follicular Lymphoma Evaluation
Index (FLEX ): a new clinical prognostic model that is superior
to existing risk scores for predicting progressionfree survival
and early treatment failure after frontline immunochemotherapy.
Am J Hematol. 2020;95(12):15031510. https://doi.org/10.1002/ajh.
25973.
16. Dave SS, Wright G, Tan B, et al. Prediction of survival in follicular
lymphoma based on molecular features of tumorinfiltrating immune
cells. N Engl J Med. 2004;351(21):21592169. https://doi.org/10.
1056/NEJMoa041869.
92
-
CASULO
17. Kridel R, Xerri L, GelasDore B, et al. The prognostic impact of
CD163positive macrophages in follicular lymphoma: a study from
the BC cancer agency and the lymphoma study association. Clin
Cancer Res. 2015;21(15):34283435. https://doi.org/10.1158/1078
0432.CCR143253.
18. Tobin JWD, Keane C, Gunawardana J, et al. Progression of disease
within 24 months in follicular lymphoma is associated with reduced
intratumoral immune infiltration. J Clin Oncol. 2019;37(34):
33003309. https://doi.org/10.1200/JCO.18.02365.
19. Pastore A, Jurinovic V, Kridel R, et al. Integration of gene mutations
in risk prognostication for patients receiving firstline immunoche-
motherapy for follicular lymphoma: a retrospective analysis of a
prospective clinical trial and validation in a populationbased regis-
try. Lancet Oncol 2015;16:11111122. https://doi.org/10.1016/
S14702045(15)001692.
20. Mondello P, Fama A, Larson MC, et al. Intrafollicular CD4+Tcells as
an independent predictor of early clinical failure in newly diagnosed
follicular lymphoma. 2019;134(suppl_1):121.
21. Huet S, Tesson B, Jais JP, et al. A geneexpression profiling score for
prediction of outcome in patients with follicular lymphoma: a
retrospective training and validation analysis in three international
cohorts. Lancet Oncol. 2018;19:549561. https://doi.org/10.1016/
S14702045(18)301025.
22. Trotman J, Luminari S, Boussetta S, et al. Prognostic value of PETCT
after firstline therapy in patients with follicular lymphoma: a pooled
analysis of central scan review in three multicentre studies. Lancet
Haematol. 2014;1(1):e17e27. https://doi.org/10.1016/S23523026
(14)700080.
23. Meignan M, Cottereau AS, Versari A, et al. Baseline metabolic tumor
volume predicts outcome in hightumorburden follicular lymphoma:
a pooled analysis of three multicenter studies. J Clin Oncol.
2016;34:3618. https://doi.org/10.1200/JCO.2016.66.9440.
24. DelfauLarue MH, van der Gucht A, Dupuis J, et al. Total metabolic
tumor volume, circulating tumor cells, cellfree DNA: distinct prog-
nostic value in follicular lymphoma. Blood Adv. 2018;2(7):807816.
https://doi.org/10.1182/bloodadvances.2017015164.
25. Morschhauser F, Tilly H, Chaidos A, et al. Tazemetostat for patients
with relapsed or refractory follicular lymphoma: an openlabel, single
arm, multicentre, phase 2 trial. Lancet Oncol. 2020;21(11):14331442.
https://doi.org/10.1016/S14702045(20)304411.
26. Jurinovic V, Passerini V, Oestergaard MZ, et al. Evaluation of the
m7FLIPI in patients with follicular lymphoma treated within the
Gallium trial: EZH2 mutation status may be a predictive marker for
differential efficacy of chemotherapy. Blood. 2019;134(suppl_1):122.
https://doi.org/10.1182/blood2019130208.
How to cite this article: Casulo C. Upfront identification of
highrisk follicular lymphoma. Hematological Oncology.
2021;39(S1):8893. https://doi.org/10.1002/hon.2852
CASULO
-
93
Received: 26 February 2021
DOI: 10.1002/hon.2853
SUPPLEMENT ARTICLE
Highrisk follicular lymphoma: Treatment options
Brad Kahl
1,2
1
Medicine, Washington University in St. Louis,
St. Louis, Missouri, USA
2
Lymphoma Program, Siteman Cancer Center,
St. Louis, Missouri, USA
Correspondence
Brad Kahl, Medicine, Washington University
in St. Louis, St. Louis, Missouri, USA.
Email: bkahl@wustl.edu
Abstract
Follicular lymphoma (FL) is the most common indolent nonHodgkin lymphoma in
the Western hemisphere. The natural history of FL appears to have been favor-
ably impacted by the introduction of rituximab. Randomized clinical trials have
demonstrated that the addition of rituximab to standard chemotherapy induction
has improved the overall survival. Maintenance rituximab strategies can improve
progressionfree survival (PFS). Obinutuzumab was superior to rituximab for PFS
in the GALLIUM study, although the benefit was small and required more drug.
Chemotherapy platforms have changed in the past decade, as bendamustine
combined with rituximab has become commonly utilized frontline strategy in
North America and parts of Europe, although there is certainly no one standard
treatment. However, several unmet needs remain, including a better ability to
identify highrisk patients at diagnosis, the development of predictive biomarkers
for targeted agents, the development of novel combinations, and strategies to
reduce the risk of transformation. A multitude of novel therapies are under
investigation in both the frontline and relapsed/refractory settings. It will be
critical to identify the most appropriate populations for new agents and to
develop validated surrogate endpoints, so that novel agents can be tested (and
adopted, if appropriate) efficiently.
KEYWORDS
follicular lymphoma, highrisk, therapy
1
|
THERAPY OF SYMPTOMATIC, HIGHTUMOR
BURDEN FOLLICULAR LYMPHOMA
The addition of rituximab to conventional chemotherapy, has
improved outcomes in follicular lymphoma (FL), including overall
and complete response rates, progressionfree survival (PFS) and
overall survival (OS). The results were remarkably consistent across
four randomized clinical trials. Clearly, rituximab added to chemo-
therapy has a therapeutic advance in FL. However, newer data
raises questions regarding the optimal antiCD20 monoclonal
antibody and the optimal chemotherapy backbone remains
unsettled.
2
|
WHICH ANTICD20 MONOCLONAL
ANTIBODY IS BEST?
The landmark GALLIUM trial compared obinutuzumab, a glyco
engineered type II antiCD20 monoclonal antibody against ritux-
imab in the frontline management of hightumor burden FL.
1
The
study randomly assigned more than 1200 patients across the globe
to either six cycles of either obinutuzumab +chemotherapy or rit-
uximab +chemotherapy, followed by maintenance antiCD20 for 2
years. Each participating center selected a chemotherapy backbone,
with the option of bendamustine, cyclophosphamide, adriamycin,
vincristine, prednisone (CHOP), or CVP. The rates of chemotherapy
94
-
Hematological Oncology. 2021;39(S1):9499. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
use were approximately 60% for bendamustine, 30% for CHOP, and
10% for CVP. The trial demonstrated a statistically significant
reduction in the risk for progression or death in the patients assigned
to obinutuzumab +chemotherapy. The hazard ratio was 0.66, which
equates to a 34% risk reduction. At the landmark of 3 years, 80% of
the obinutuzumab +chemotherapy patients were in remission versus
73% of the rituximab +chemotherapy patients. No OS differences
were observed. The obinutuzumabtreated patients had slightly more
infusion reactions, episodes of cytopenia, and infections than the
rituximabtreated patients. So is obinutuzumab a superior antiCD20
monoclonal antibody? Certainly, the outcome of PFS was certainly
better in the obinutuzumabtreated patients. However, the results
must be interpreted in the context of the dose and schedule. Patients
received higher doses of obinutuzumab and more doses of obinutu-
zumab. A patient who is 2 m
2
would receive 36% more obinutuzumab
than rituximab on this study. Note the 35% risk reduction is very
similar to the 35% difference in dosing. As a result of the design
issues with GALLIUM, it is impossible to know if obinutuzumab is
truly a better monoclonal antibody than rituximab in FL or whether
the results observed are simply of function of dosing differences. As a
result, obinutuzumab has not been universally accepted as the new
standard of care and some investigators continue to administer
rituximab.
3
|
WHICH CHEMOTHERAPY IS THE BEST?
Before the introduction of bendamustine, the most commonly used
regimens in the United States were RCHOP (rituximab, cyclophos-
phamide, vincristine, prednisone) (60%), RCVP (rituximab, cyclo-
phosphamide, prednisone) (27%), and Rfludarabinebased regimen
(13%). A randomized comparison of these regimens indicated R
CHOP had the best riskbenefit profile of the three, as it was more
active than RCVP and less toxic than RFM.
2
The alkylating agent bendamustine has gained widespread,
although not universal, adoption as the chemotherapy platform of
choice in FL. A phase III trial from the StIL group comparing bend-
amustine, rituximab (BR) to RCHOP demonstrated BR had better
efficacy and reduced toxicity. In this multicenter phase III study, 549
patients with hightumor burden indolent non Hodgkin lymphoma
(NHL) and mantle cell lymphoma (MCL) (median age 64 years) were
randomized to receive bendamustine 90 mg/m
2
on days 1 and 2, with
rituximab 375 mg/m
2
on day 1, every 28 days (the BR group) or to
receive standard RCHOP chemotherapy every 21 days. The overall
response rates (ORRs) were similar in the BR versus RCHOP groups
(92.7% vs. 91.3%, respectively), but the complete response (CR) rate
was significantly higher in the BR group (39.8%) compared with the
RCHOP group (30.0%) (p=0.03). When evaluating just the FL pa-
tients, with a median followup of 45 months, the median PFS was
significantly longer after BR compared with RCHOP (median PFS,
not reached vs. 40.9 months, p=0.007). OS did not differ between
both groups. There were less hematologic toxicity, alopecia, in-
fections, peripheral neuropathy, and stomatitis with BR. Drug
associated erythematous skin reactions were seen more frequently
in the BR group. These data suggest that BR is a better option for
untreated hightumor burden FL. Nineyear updates from this trial
were presented at the 2017 American Society of Clinical Oncology
Meetings.
3
The longterm followup indicated the PFS benefit per-
sisted over time but no OS benefit emerged.
A confirmatory randomized phase III trial (BRIGHT study) was
conducted in North America. Previously untreated indolent NHL
patients with hightumor burden were randomized to BR or R
CHOP/RCVP. Control arm patients were identified as an RCHOP or
RCVP candidate prior to randomization. The primary endpoint was
to show noninferiority of BR in the CR rate. Seventy percent of the
447 enrolled patients had FL, and in these patients, BR therapy was
found to be noninferior to the RCHOP/RCVP control arm for CR
rate (30% vs. 25%) and the ORR (99% vs. 94%). Sideeffect profiles
were distinct, with more GI toxicity and rash with BR, and more
neuropathy and alopecia with RCHOP/RCVP. Longterm followup
was reported in 2019.
4
When one separates out the mantle cell
lymphoma patients and limits the analysis to indolent lymphoma, BR
was not statistically superior to RCHOP/RCVP for PFS, although
there was trend toward improvement in PFS with BR with a hazard
ratio of 0.70 (95% CI: 0.49–1.01). Although, the BRIGHT data do not
exactly replicate the StIL data for BR, they do suggest that BR re-
mains an attractive alternative to RCHOP or RCVP in FL.
4
|
WHO SHOULD RECEIVE MAINTENANCE ANTI
CD20 THERAPY?
The question of whether to administer maintenance rituximab after
frontline Rchemotherapy was addressed in the phase III PRIMA trial.
The study evaluated the efficacy and safety profile of maintenance
rituximab in newly diagnosed FL patients who responded to initial
treatment with rituximab plus chemotherapy. Induction treatment
was selected by center; RCHOP (75%), RCVP (22%), or RFCM
(3%). Patients were randomized to either observation or a single
dose of rituximab every 2 months for 2 years. At a median followup
of 36 months from randomization, the 2year PFS in the maintenance
rituximab arm was 75% versus 58% in the observation arm
(p<0.0001). The beneficial effect of maintenance rituximab was seen
irrespective of the induction chemotherapy backbone and in both CR
and partial remission (PR) patients. Grade 3–4 adverse events were
slightly higher in the maintenance rituximab arm (24% vs. 17%).
Longterm followup from this dataset was published recently.
5
With
a median followup of 9 years, the 10year PFS estimates were 51.1%
in the rituximab maintenance arm and 35.0% in the no maintenance
arm. No difference in OS was observed, with both groups exhibiting
an OS of 80% at 10 years. No new safety signals emerged. Given the
lack of OS benefit, the decision regarding the use of maintenance
rituximab can be individualized, but a 50% likelihood of remaining
disease free at 10 years is highly appealing and begs the question of
whether a proportion of FL patients are actually being cured with
frontline therapy.
KAHL
-
95
5
|
WHAT ABOUT MAINTENANCE ANTICD20
THERAPY AFTER BENDAMUSTINE?
A highly interesting and informative subgroup analysis on chemo-
therapy selection arose from the GALLIUM study.
6
The chemo-
therapy choice (bendamustine vs. CHOP vs. CVP) was center
dependent and not the result of a randomization process. As a result,
these data must be interpreted with caution, but several aspects are
noteworthy. The 3year PFS rates were highest in the bendamustine
group and lowest in the CVP group. The risk of grade 3–5 adverse
events was higher with CHOP than with CVP or bendamustine, but
most of these events were transient cytopenias. Grade 3–5 infections
were more likely with bendamustine than with CHOP or CVP and
fatal adverse event (AEs) were more common with bendamustine
than with CHOP or CVP. The difference in infections was particularly
notable during the maintenance and followup phase. A potential
explanation is the significant difference in CD4+Tcell suppression
following treatment. Patients treated with bendamustine experi-
enced profound decreases in CD4+Tcell counts during bendamus-
tine treatment, with the nadir occurring at end of induction and
failing to recover to baseline even 3 years after treatment, whereas
the Tcell counts were impacted negligibly by CHOP and CVP ther-
apy. It should be noted that the bendamustinetreated patients were
slightly older on average than the CHOPtreated patients, with more
patients over age 70 and 80. However, the age distribution of the
bendamustine and CVPtreated patient was quite similar. Given the
significant Tcell depletion experienced after bendamustinebased
treatment, it is quite possible that maintenance therapy with anti
CD20 therapy (and its associate prolonged Bcell depletion) tips
the risk–benefit ratio unfavorably in FL. Additional data on the risks
versus benefits of maintenance rituximab should come from the Stil
NHL72008 Maintain trial, which prospectively compares 2 versus 4
years of maintenance rituximab following BR induction therapy.
Initial results suggested no major benefit for 4 years over 2 years and
an acceptable safety profile.
7
Attempts to improve the outcomes standard immunochemo-
therapy with the addition of novel agents have been unsuccessful to
date. For example, E2408 was a threearm randomized phase II
clinical trial, using BR followed by rituximab maintenance as the
control arm, added the proteasome inhibitor bortezomib to standard
BR induction therapy in one of the experimental arms (BVR with R
maintenance), and added the immunomodulatory agent lenalidomide
to maintenance rituximab in the third arm (BR with lenalidomide,
rituximab (LR) maintenance). The 3year PFS was similar in all three
arms, suggesting no benefit for these interventions.
8
6
|
IS SINGLE AGENT RITUXIMAB SUFFICIENT IN
THIS PATIENT POPULATION?
Most of the data examining the role for single agent rituximab in the
frontline management of FL has been generated in patients who
were asymptomatic and had lowtumor burden by GELF criteria
(see RESORT Trial from Kahl et al., or UK Trial from Ardeshna et al.).
For patients with highrisk FL, due to hightumor burden, there is no
question that single agent rituximab is not as potent as rituximab plus
chemotherapy. Having stated that, there may be situations where it is
reasonable to consider single agent rituximab. Patients may be
elderly and frail, or have significant comorbidities, and the risk of
incorporating chemotherapy may be unacceptably high in certain
circumstances. Or patients may simply refuse treatment with cyto-
toxic chemotherapy for personal reasons. Single agent rituximab can
be adequately efficacious as was demonstrated by Ghielimini et al. in
the SAKK 35/98 trial, where patients were not required to have low
tumor burden. The single agent ORR was 67%. Previously untreated
patients receiving a maintenance strategy (n=20) enjoyed a 45%
probability is remaining diseasefree at 5 years, suggesting this
strategy is perfectly reasonable for selected patients. Prior work
from Ghielmini suggests and single agent rituximab loses efficacy
when lymph nodes exceed 5 cm in size and work from Witzig et al.
showed single agent rituximab is quite ineffective in frontline FL if
the lactate dehydrogenase (LDH) is elevated. To summarize, single
agent rituximab can be considered a reasonable option in hightumor
burden FL, particularly in elderly or frail patients but clinicians should
be aware that the efficacy appears to be diminished if the tumor
burden is too high or if there is evidence for high proliferation.
7
|
ARE THERE NOVEL STRATEGIES WORTHY OF
FRONTLINE CONSIDERATION?
A novel strategy, combining the immunomodulatory agent lenalido-
mide with rituximab, for the initial management of FL was first re-
ported by investigators from the MD Anderson Cancer Center. The
promising results led to the launch of the RELEVANCE trial.
9
A total
of 1030 patients with previously untreated FL were randomly
assigned to the standard arm of rituximab plus chemotherapy (in-
vestigators choice of RCHOP, BR, or RCVP) or the experimental
arm of lenalidomide plus rituximab. For the first six cycles, lenalido-
mide was administered orally at 20 mg/day for 3 out of every 4 weeks
of each 28day cycle and rituximab was given at 375 mg/m
2
Intra-
venous (IV) on day 1 of each cycle. Patients responding after cycle 6
could continue on therapy for up to 12 more cycles, receiving lena-
lidomide at a dose of 10 mg and rituximab every 8 weeks. The efficacy
between the two strategies was similar in terms of ORR, complete
response rate, PFS and OS. The 3year PFS was 77% in the lenali-
domide rituximab arm compared to 78% in the Rchemotherapy arm.
There was more grade 3–4 neutropenia and febrile neutropenia in the
chemotherapy arms and more rash and cutaneous reactions in the
lenalidomide arm. Despite the excellent results obtained for lenali-
domide–rituximab, it was technically a “negative” study since it was
designed (and failed) to show superiority for the novel combination,
and it has not received a frontline indication in the US or EU. The MD
Anderson group has completed a phase II study of obinutuzumab–
lenalidomide, with promising initial results, exhibiting an ORR of
100% and CR rate of 75% in 57 patients with hightumor burden FL.
10
96
-
KAHL
8
|
WHAT ABOUT UNIQUE PATHOBIOLOGIC
SITUATIONS?
There are several potential scenarios that warrant special mention. If
a patients’ biopsy does not demonstrated transformation but the
clinical presentation creates a high level of suspicion for trans-
formation, such as significant B symptoms, hypercalcemia, signifi-
cantly elevated LDH, very high standard uptake values on positron
emission tomography (PET) imaging, is prudent to select
anthracyclinebased immunochemotherapy. If the patients’ diagnostic
biopsy demonstrates grade 3B FL, the preponderance of data sug-
gests these patients are best managed with anthracyclinebased
immunochemotherapy. If the patients’ diagnostic biopsy demon-
strates grade 3A FL, then these patients can be reasonably managed
like typically grade 1–2 FL. Finally, the entity of duodenal FL appears
to have a unique underlying biology and natural history that is more
indolent than typical FL. These patients can often be very successfully
managed with prolonged periods of no active treatment or at most,
singleagent rituximab, and rarely should chemotherapy be required.
9
|
SUMMARY OF FRONTLINE TREATMENT
CONSIDERATIONS
The last 20 years of research into the management of hightumor
burden FL has cemented immunochemotherapy as the standard
approach. However, there is no one universally standard immu-
nochemotherapy regimen. In fact, there is quite a bit of geographic
variation in the standards. The most commonly utilized options
include either obinutuzumab or rituximab combined with either
bendamustine or CHOP chemotherapy. Given the lack of OS benefit
for maintenance therapy with an antiCD20, it can be considered
optional. Maintenance antiCD20 therapy has a fairly profound
impact on PFS after RCHOP, and thus is often utilized. On the other
hand, maintenance antiCD20 appears to have much less benefit
after BR therapy, and the GALLIUM study suggests it may do more
harm than good. A summary of outcomes from selected frontline
studies is included in Table 1.
9.1
|
Therapy for relapsed and refractory follicular
lymphoma
Multiple options exist for the treatment of patients who have failed
firstline therapy, and the decision of which therapy to use depends on
a number of factors, including the prior treatment utilized, duration of
prior response, patient age, comorbid illnesses, and goals of therapy.
Duration of prior responses has consistently been shown to be a
powerful predictor of future outcomes. Patients who experience dis-
ease progression within 24 months (POD24) of frontline immu-
nochemotherapy have been shown to have a 50% overall survival at 5
years, compared with over 90% survival for patients who do not
experience POD24. How to best manage POD24 patients remains
unclear. Retrospective data from stem cell transplant registries have
suggested superior outcomes for transplant strategies in this popu-
lation, and certainly should be considered in young, fit patients. One
word of caution, however, investigators from the British Columbia
Cancer Agency have reported that up to 50% of the POD24 population
TABLE 1Efficacy data in frontline
hightumor burden follicular lymphoma Trial NInduction Maintenance 3year PFS (%)
PRIMA 513 RCHOP/RCVP/RFCM None 58
505 RCHOP/RCVP/RFCM Rituximab 75
STiL 140 RCHOP None 55
139 BR None 75
BRIGHT 186 RCHOP/RCVP Rituximab in 45% 75
187 BR Rituximab in 43% 80
GALLIUM 341 BR Rituximab 81
345 BO Obinutuzumab 85
203 RCHOP Rituximab 77
196 OCHOP Obinutuzumab 82
57 RCVP Rituximab 77
60 OCVP Obinutuzumab 77
E2408 65 BR Rituximab 77
99 BVR Rituximab 82
125 BR Rlenalidomide 76
RELEVANCE 517 RCHOP/BR/RCVP Rituximab 78
513 Rlenalidomide Rlenalidomide 77
KAHL
-
97
may harbor transformed disease suggesting it is imperative to inves-
tigate this possibility before choosing additional therapy in a POD24
patient.
Besides stem cell transplantation, a variety of nonintensive
strategies exist for R/R FL. Certainly, one can administer repeat
course of immunochemotherapy (e.g., patients relapsing after BR can
receive 0CHOP). The increasing menu of novel agents is providing
additional options for management. There are now four PI3 kinase
inhibitors available (idelalisib, duvelisib, copanlisib, and umbrali-
sib).
11–13
Response rates for all three agents tend to run in the 45%–
60% range and average durability is typically around 11 months. The
oral PI3k inhibitors carry the risk of pneumonitis and colitis, although
the risk with umbralisib may be less. Copanlisib does not seem to
have these risks although transient hypertension and hyperglycemia
around administration must be monitored and addressed. More
promising than the PI3k inhibitors is the combination of lenalidomide
and rituximab, as demonstrated in the AUGMENT study.
14
The ORR
for the combination was 78% and the median PFS was 39 months,
compared to 53% and 14 months for rituximab plus placebo. Toxic-
ities were increased compared to the rituximab control arm, most
notably for infection (63% vs. 49%), neutropenia (58% vs. 23%), and
cutaneous reactions (32% vs. 12%). Given the marked improvement
in efficacy, the increased toxicity profile is quite acceptable. Finally, a
firstinclass EZH2 inhibitor, tazemetostat, received FDA approval in
R/R FL in the summer of 2020. Activating mutations of EZH2, an
epigenetic regulator, appear to play significant roles in the patho-
genesis of FL. Accordingly, agents designed to suppress EZH2 enzy-
matic activity are being developed. Tazemetostat 15 is an oral EZH2
inhibitor and was tested in 99 patients with R/R FL, 45 of whom
had mutated EZH2 genes and 54 with wildtype EZH2.
15
In the
EZH2mutated patients, the ORR was 69% and the median PFS was
13.8 months. In the EZH2 wildtype patients, the ORR was 35% and
the median PFS was 11.1 months. The safety profile appears very
good, with negligible grade 3–4 toxicities. The regulatory approval
includes both mutated and wildtype patient populations. Table 2
summarizes data for the US FDAapproved targeted agents in R/R FL.
There are several promising investigational immunotherapy ap-
proaches in development in relapsed FL. Targeting the “don't eat me”
CD47 antigen, expressed on many tumor types and preventing
macrophage engulfment, with monoclonal antibodies against CD47
appears promising in early studies. Magrolimab demonstrated an
ORR of 61% and CR rate of 24%, as reported by Advani et al. and the
2019 ICML meeting. The bispecific monoclonocal antibody mosune-
tuzumab targets both CD20 and CD3 and is a potent Tcell engager.
An ongoing trial has demonstrated an ORR of 68% with an impres-
sive CR rate of 52% in R/R FL. Finally, and perhaps most encouraging,
is recent data generated testing CART therapy in R/R FL. Axi-
cabtagene ciloleucel (axicel) demonstrated on ORR of 94% with an
CR rate of 80% in a population of R/R FL patients. With a median
followup of 17 months, the 12month PFS rate is 77%. More study
and longer followup are needed for these immunooncology agents,
but the early data are very promising and summarized in Table 3.
10
|
SUMMARY
Outcomes are generally very good in FL with median OS exceeding
15 years. As a result, it is sometimes deprioritized as a cancer in need
of therapeutic advances. Certainly, there are FL patients who never
require therapy or who only require one line of therapy and can be
TABLE 2Efficacy data for approved, targeted agents in R/R follicular lymphoma
Agent(s) Class NORR (%) complete response rate (CRR) (%) median progression free survival (mPFS) (m)
Lenalidomide–rituximab Immunomodulatory 147 78 34 39
Rituximab antiCD20 MoAb 148 53 18 14
Idelalisib PI3k inhibitor 72 60 15 11
Duvelisib PI3k inhibitor 83 48 2 10
Copanlisib PI3k inhibitor 104 60 14 11
Umbralisib PI3k inhibitor 117 45 5 11
Tazemetostat (mutated) EZH2 inhibitor 45 69 13 14
Tazemetostat (wildtype) EZH2 inhibitor 54 35 4 11
Abbreviation: ORR, overall response rate.
TABLE 3Investigational
immunotherapy agents with promising
activity in R/R follicular lymphoma
Agent Class NORR (%) CRR (%) Presented
Magrolimab AntiCD47 MoAb 41 61 24 Advani, Lugano 2019
Mosunetuzumab Bispecific MoAb 62 68 52 Assouline, ASH 2020
Axicel CART 84 94 80 Jacobson, ASH 2020
Abbreviation: ORR, overall response rate.
98
-
KAHL
considered functional cures. On the other hand, subsets of FL pa-
tients are at high risk of death from their disease. These include
patients with highrisk m7FLIPI scores, patients who experience
recurrence within 2 years of Rchemotherapy, patients who experi-
ence histologic transformation, and FL patients under the age of 60
at diagnosis. Future research should (a) seek to identify prognostic
biomarkers capable of identifying highrisk patients at diagnosis; (b)
continue to develop targeted therapies (with predictive biomarkers);
(c) test interventions designed to reduce the risk for histologic
transformation; and (d) seek to reliably cure FL.
CONFLICT OF INTEREST
Dr. Kahl reports consulting fees from Abbvie, Acerta, Celgene,
Genentech, Roche, Pharmacyclics, Gilead, Bayer, AstraZeneca,
Beigene. Dr. Kahl reports research funding from Acerta, Celgene,
Genentech, Beigene.
DATA AVAILABILITY STATEMENT
All data provided in this manuscript is publicly available.
PEER REVIEW
The peer review history for this article is available at https://publons.
com/publon/10.1002/hon.2853.
REFERENCES
1. Marcus R, Davies A, Ando K, et al. Obinutuzumab for the firstline
treatment of follicular lymphoma. N Engl J Med. 2017;377(14):1331
1344. https://doi.org/10.1056/NEJMoa1614598.
2. Federico M, Luminari S, Dondi A, et al. RCVP versus RCHOP
versus RFM for the initial treatment of patients with advanced
stage follicular lymphoma: results of the FOLL05 trial conducted
by the Fondazione Italiana Linfomi. J Clin Oncol. 2013;31(12):1506
1513. https://doi.org/10.1200/JCO.2012.45.0866.
3. Rummel MJ, Maschmeyer G, Ganser A, et al. Bendamustine plus
rituximab (BR) versus CHOP plus rituximab (CHOPR) as firstline
treatment in patients with indolent lymphomas: nineyear updated
results from the StiL NHL1 study. J Clin Oncol. 2017;35(15_suppl):
75017501. https://doi.org/10.1200/JCO.2017.35.15_suppl.7501.
4. Flinn IW, van der Jagt R, Kahl B, et al. Firstline treatment of pa-
tients with indolent nonHodgkin lymphoma or mantlecell lym-
phoma with bendamustine plus rituximab versus RCHOP or RCVP:
results of the BRIGHT 5year followup study. J Clin Oncol.
2019;37(12):984991. https://doi.org/10.1200/JCO.18.00605.
5. Bachy E, Seymour JF, Feugier P, et al. Sustained progressionfree
survival benefit of rituximab maintenance in patients with follic-
ular lymphoma: longterm results of the PRIMA study. J Clin Oncol.
2019. JCO.19.01073. https://doi.org/10.1200/jco.19.01073.
6. Hiddemann W, Barbui AM, Canales MA, et al. Immunochemotherapy
with obinutuzumab or rituximab for previously untreated follicular
lymphoma in the GALLIUM study: influence of chemotherapy on
efficacy and safety. J Clin Oncol. 2018;36(23):23952404. https://doi.
org/10.1200/JCO.2017.76.8960.
7. Rummel MJ, Buske C, Hertenstein B, et al. Four versus two
years of rituximab maintenance (Rmaintenance) following bend-
amustine plus rituximab (BR): initial results of a prospective,
randomized multicenter phase 3 study in firstline follicular lym-
phoma (the StiL NHL72008 MAINTAIN study). Blood. 2017;
130(suppl 1):483483. https://doi.org/10.1182/blood.V130.Suppl_1.
483.483.
8. Evens AM, Hong F, Habermann TM, et al. A threearm randomized
phase II study of bendamustine/rituximab with bortezomib induc-
tion or lenalidomide continuation in untreated follicular lymphoma:
ECOGACRIN E2408. Clin Cancer Res. 2020;26(17):44684477.
https://doi.org/10.1158/10780432.CCR201345.
9. Morschhauser F, Fowler NH, Feugier P, et al. Rituximab plus lena-
lidomide in advanced untreated follicular lymphoma. N. Engl J Med.
2018;379(10):934947. https://doi.org/10.1056/NEJMoa1805104.
10. Fowler NH, Nastoupil LJ, Chin C, et al. A phase I/II study of lenali-
domide plus obinutuzumab in relapsed indolent lymphoma. Blood.
2019;134(suppl 1):348348. https://doi.org/10.1182/blood2019
131045.
11. Gopal AK, Kahl BS, de Vos S, et al. PI3Kdelta inhibition by idelalisib
in patients with relapsed indolent lymphoma. N Engl J Med.
2014;370(11):10081018. https://doi.org/10.1056/NEJMoa131
4583.
12. Flinn IW, Miller CB, Ardeshna KM, et al. DYNAMO: a phase II study
of duvelisib (IPI145) in patients with refractory indolent non
Hodgkin lymphoma. J Clin Oncol. 2019;37(11):912922. https://doi.
org/10.1200/JCO.18.00915.
13. Dreyling M, Santoro A, Mollica L, et al. Phosphatidylinositol 3kinase
inhibition by copanlisib in relapsed or refractory indolent lymphoma.
J Clin Oncol. 2017;35(35):38983905. https://doi.org/10.1200/JCO.
2017.75.4648.
14. Leonard JP, Trneny M, Izutsu K, et al. AUGMENT: a phase III study
of lenalidomide plus rituximab versus placebo plus rituximab in
relapsed or refractory indolent lymphoma. J Clin Oncol.
2019;37(14):11881199. https://doi.org/10.1200/JCO.19.00010.
15. Morschhauser F, Tilly H, Chaidos A, et al. Tazemetostat for pa-
tients with relapsed or refractory follicular lymphoma: an open
label, singlearm, multicentre, phase 2 trial. Lancet Oncol.
2020;21(11):14331442. https://doi.org/10.1016/S14702045(20)
304411.
How to cite this article: Kahl B. Highrisk follicular lymphoma:
Treatment options. Hematological Oncology.
2021;39(S1):9499. https://doi.org/10.1002/hon.2853
KAHL
-
99
Received: 15 February 2021
DOI: 10.1002/hon.2845
SUPPLEMENT ARTICLE
Allogeneic stem cell transplant in nonHodgkin lymphomas:
Still an indication?
Peter Dreger
Department of Medicine V, University of
Heidelberg, Heidelberg, Germany
Correspondence
Peter Dreger, Department Medicine V,
University of Heidelberg, Im Neuenheimer
Feld 410, 69120 Heidelberg, Germany.
Email: peter.dreger@med.uni-heidelberg.de
Abstract
Allogeneic hematopoietic cell transplantation (alloHCT) used to play a defined role in
the treatment of nonHodgkin lymphoma (NHL). With the advent of modern targeted
molecular therapies and immunotherapies, treatment standards at least for Bcell
lymphoma have undergone significant changes, thereby questioning the traditional
role of alloHCT in these diseases. This paper attempts to describe the current place
and the perspectives of alloHCT in the rapidly evolving treatment landscape of NHL.
KEYWORDS
allogeneic transplantation, CART cells, lymphoma, NHL
1
|
INTRODUCTION
Entering the clinical stage more than 50 years ago, allogeneic he-
matopoietic cell transplantation (alloHCT) was the first immuno-
therapy successfully applied to patients, and can be considered as the
ancestor of modern cellular immunotherapy.
1
Despite its inherent
drawbacks of significant nonrelapse mortality and morbidity due to
graftversushost disease, until recently, alloHCT has been playing a
defined role in the management algorithms of the main NHL subtypes
(i.e., diffuse large Bcell lymphoma [DLBCL]; follicular lymphoma [FL];
mantle cell lymphoma [MCL]; and peripheral Tcell lymphoma
[PTCL]), mostly in the salvage setting. With the advent of modern
targeted molecular therapies and immunotherapies, treatment stan-
dards at least for Bcell lymphoma have undergone substantial
changes, thereby questioning the traditional role of alloHCT in these
diseases. This paper attempts to describe the current place and the
perspectives of alloHCT in the rapidly evolving treatment landscapes
of DLBCL, FL, MCL, and PTCL.
2
|
DIFFUSE LARGE BCELL LYMPHOMA
As in almost all neoplastic indications where it is effective, the basis
of alloHCT in DLBCL is graftversuslymphoma activity (GVL).
Circumstantial evidence for GVL efficacy in DLBCL can be derived
from the effectiveness of immunomodulation for preventing or
treating posttransplant relapse, and from observations showing that
nonmyeloablative alloHCT can provide longterm disease control in
patients having failed autologous hematopoietic cell transplantation
(autoHCT). However, compared to the other main NHL entities,
DLBCL appears to be less GVLsensitive resulting in longterm
progressionfree survival rates of 30%–40%, and even less if calcu-
lated by intenttotreat.
2–4
While the standard indication for alloHCT in DLBCL used to
be chemosensitive disease following failure of autoHCT, chimeric
antigen receptorengineered Tcells (CARTs) have become the
preferred cellular immunotherapy in this setting.
3
A preliminary
intenttotreat comparison of alloHCT versus CARTs in large Bcell
lymphoma (LBCL) suggested that in patients having failed at least
two lines of systemic therapy, survival tended to be better with
the CART versus the alloHCT approach.
5
One limitation of this
study was the short followup of 10 months in the CART group.
Figure 1shows an update of this comparison with a median
followup of 15 (9–24) months, suggesting that the favorable trend
for CARTs was maintained, both measured from treatment indi-
cation and start of cellular therapy.
In conclusion, for the time being, alloHCT remains an indication
for advanced DLBCL when CARTs have failed or are not feasible.
3
This is an open access article under the terms of the Creative Commons AttributionNonCommercial License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited and is not used for commercial purposes.
© 2021 The Authors. Hematological Oncology published by John Wiley & Sons Ltd.
100
-
Hematological Oncology. 2021;39(S1):100103. wileyonlinelibrary.com/journal/hon
However, it is clear that emerging innovations, such as bispecific
antibodies and moving CARTs to the second line, have the potential
for further modifying the importance of alloHCT for rescuing patients
with relapsed/refractory DLBCL.
3
|
FOLLICULAR LYMPHOMA
Although FL appears to be the most GVLsensitive disease among all
alloHCT indications currently considered as standard, with a 5year
relapse risk less than 20% in all major studies published,
2,6
the
number of allogeneic transplantations for this subtype is decreasing
since several years, making FL the least frequent allo indication of the
four entities discussed in this review. In 2018, the numbers of allo-
transplants registered with the European Society for Blood and
Marrow Transplantation (EBMT) for PTCL, DLBCL, MCL, and FL were
401, 320, 161, and 144, respectively (EBMT data on file, PROMISE
download 19 February 2020). This has to do with the indolent course
of the disease, the high efficacy of standard firstline treatment with
chemoimmunotherapies and CD20 antibody maintenance, and the
availability of a broad effective toolkit for salvage treatment,
including revlimid, targeted therapies, and also autoHCT. Even in
highrisk disease, defined by failure of firstline treatment within
2 years (“POD24”), there is no proven benefit of alloHCT over
autoHCT.
7
In contrast to DLBCL and MCL, there is no CART therapy
approved for FL available to date, and due to the indolent character
of the disease the benefit of CART approaches in terms of durable
lymphoma control will be more difficult to assess. Nevertheless, the
development of CD19directed CARTs and other immunotherapies
for indolent lymphoma is already quite advanced, suggesting that the
alloHCT indication will further narrow in the near future.
For today, alloHCT is still a potentially curative option for those
patients with FL who are resistant to less aggressive approaches, that
is, who relapse early after salvage autoHCT or a similarly intensive
regimen, and for patients with emerging exhaustion of hematopoie-
sis, or incipient myelodysplasia.
4
|
MANTLE CELL LYMPHOMA
Efficacy of donor lymphocyte infusions (DLI) and plateaus in the
relapse curves after reduced intensity conditioning (RIC) suggest that
there is a biologically relevant contribution of GVL also in MCL.
8,9
Given the poor prognosis of MCL recurring after stateoftheart
intensive firstline treatment with consolidating autoHCT and ritux-
imab maintenance, the traditional place of (RIC) alloHCT has been
consolidation of secondline responses. With the introduction of
Bruton's tyrosine kinase inhibitors (BTKi) as standard of care salvage
therapy in MCL, and the recent approval of the CART product
brexucabtagene autoleucel for relapsed/refractory MCL, the place of
alloHCT in the MCL management algorithm needs to be reevaluated.
A recent international consensus project recommended considering
alloHCT in MCL only if CARTs have failed or are not feasible
(Hamadani et al., manuscript in preparation). This would mean that
alloHCT comes into play only on the fourth place after standard in-
duction, BTKi, and CARTs. However, in areas where CARTs are not
available, considering alloHCT already for consolidation of second
line responses to BTKi might be worthwhile in highrisk patients,
such as those without a complete response to BTKi or early failure
after standard induction.
10
Again, the expected advent of novel
molecular agents and immunotherapeutics in the clinical routine of
MCL management, such as venetoclax, enhanced CD19directed
antibodies, and nextgeneration phosphatidylinositol 3kinase delta
inhibitors has the potential to further differentiate the alloHCT
indication in MCL in the near future.
5
|
PERIPHERAL TCELL LYMPHOMA
In contrast to Bcell lymphoma, there have been no major therapeutic
improvements for PTCL in the last decades, with the exception of
anaplastic large cell lymphoma (ALCL). Although the value of autoHCT
consolidation is ambiguous, accepted standard treatment for the three
predominant nodal PTCL subsets (i.e., ALKnegative ALCL, PTCL not
other specified [PTCLNOS], and angioimmunoblastic Tcell lymphoma
FIGURE 1 Overall survival of patients with diffuse large Bcell lymphoma beyond the second line by cellular immunotherapy intended
(Heidelberg, 2005–2020, N=80). (A) Survival from indication; (B) Survival from administration of cellular immunotherapy. CAR, chimeric
antigen receptor Tcells; HCT, allogeneic hematopoietic cell transplantation; HCT>2L, HCT intended after having failed two or more lines
DREGER
-
101
[AITL]) consists in CHOPlike induction followed by highdose inten-
sification.
11
With this strategy, 5years progressionfree survival rates
of 35%–45% can be expected.
12
Of particular concern in PTCL is the
high rate of primary refractoriness which can affect up to one third of
the patients.
Similar to FL and MCL, PTCL appears to be quite susceptible to
GVL effects as illustrated by survival plateaus around 50% after
alloHCT across numerous studies and efficacy of DLI,
2,4,8,12,13
prompting the exploration of allotransplantation as firstline
consolidation. A large randomized trial comparing alloHCT with
autoHCT in the firstline setting, however, failed to show a benefit for
the alloHCT strategy, largely because the lower relapse risk associ-
ated with alloHCT was neutralized by excess nonrelapse mortality.
14
Thus, except for selected orphan PTCL subsets such as hepatosplenic
Tcell lymphoma, firstline alloHCT should not be performed outside
of clinical trials.
In contrast, alloHCT is the preferred option in relapsed/re-
fractory PTCL, ideally after having achieved a state of controlled
disease prior to transplant.
11,12
This is because of the lack of thera-
peutic alternatives with curative perspective. In patients not having
undergone autoHCT during firstline treatment, also auto-
transplantation may be considered though appearing inferior to
alloHCT in intenttotreat comparisons.
13
Similar to the other lym-
phoma subsets described in this paper, reduced intensity condition-
ing provides outcomes in PTCL that are at least similar to that
observed after myeloablative conditioning, and haploidentical do-
nors seem to be a valuable alternative if matched related or unre-
lated donors are not available.
12
6
|
CONCLUSIONS
Although substantial therapeutic innovations in particular for Bcell
lymphoma have entered the clinical stage recently, or are at the
doorstep, immunotherapy by alloHCT remains an effective and
potentially curative option for settings where the medical need un-
met by traditional chemotherapy can also not be covered by novel
therapeutics. This gap might be bigger than believed as some
promises of novel agents given in phase2 studies have not been fully
kept in the real world.
15
Suggestions for current indications for
alloHCT are summarized in Table 1. If alloHCT is taken into account
according to these suggestions, it has to be kept in mind that the
window of opportunity for a successful outcome of transplantation is
largest before tumor refractoriness and performance status deteri-
oration have developed through serial palliative or experimental
treatment attempts.
ACKNOWLEDGMENT
I thank Herve Finel for providing the EBMT registry data used here.
CONFLICT OF INTEREST
P. Dreger: consultancy for AbbVie, AstraZeneca, Gilead, Janssen,
Novartis, Riemser, Roche; speakers bureau for AbbVie, AstraZeneca,
Gilead, Novartis, Riemser, Roche; research support from Riemser.
REFERENCES
1. Mathe G, Amiel JL, Schwarzenberg L, et al. Successful allogeneic bone
marrow transplantation in man: chimerism, induced specific toler-
ance, and possible antileukemic effects. Blood. 1965;25(2):179196.
2. UrbanoIspizua A, Pavletic SZ, Flowers ME, et al. The impact of
graftversushost disease on the relapse rate in patients with
lymphoma depends on the histological subtype and the intensity
of the conditioning regimen. Biol Blood Marrow Transpl. 2015;
21(10):17461753.
3. Dreger P, Fenske TS, Montoto S, Pasquini MC, Sureda A, Hamadani
M. Cellular immunotherapy for refractory diffuse large B cell lym-
phoma in the chimeric antigen receptorengineered T cell era: still a
role for allogeneic transplantation? Biol Blood Marrow Transpl.
2020;26(4):e77e85.
4. Selberg L, Stadtherr P, Dietrich S, et al. The impact of allogeneic
hematopoietic cell transplantation on the mortality of poorrisk non
Hodgkin lymphoma: an intenttotransplant analysis. Bone Marrow
Transpl. 2021;56(1):3037.
5. Dreger P, Dietrich S, Schubert ML, et al. CAR T cells or allogeneic
transplantation as standard of care for advanced large Bcell lymphoma:
an intenttotreat comparison. Blood Adv. 2020;4(24):61576168.
6. Sureda A, Zhang MJ, Dreger P, et al. Allogeneic hematopoietic
stem cell transplantation for relapsed follicular lymphoma: a
combined analysis on behalf of the Lymphoma Working Party of
the EBMT and the Lymphoma Committee of the CIBMTR. Cancer.
2018;124(8):17331742.
7. Smith SM, Godfrey J, Ahn KW, et al. Autologous transplantation versus
allogeneic transplantation in patients with follicular lymphoma expe-
riencing early treatment failure. Cancer. 2018;124(12):25412551.
8. Robinson S, Boumendil A, Finel H, et al. Donor lymphocyte infusions
induce durable responses in patients with follicualr, mantle and T
cell lymphomas relapsing after an alloSCT. An EBMTLWP study.
Hematol Oncol. 2020;37(Suppl 2):316317. Ref Type: Abstract.
TABLE 1Select lymphoma alloHCT
standard indications at a glance
Large Bcell lymphoma: 4
th
line after failure of induction, autoHCT attempt, and CAR Tcell
therapy.
Follicular lymphoma: early relapse after salvage autoHCT or a similarly intensive regimen, and
emerging exhaustion of hematopoiesis/incipient myelodysplasia.
Mantle cell lymphoma: 4
th
line after failure of induction, BTKi, and CAR Tcell therapy.
Peripheral Tcell lymphoma: relapsed/refractory disease.
Abbreviations: alloHCT, allogeneic hematopoietic cell transplantation; autoHCT, autologous
hematopoietic cell transplantation; BTKi, Bruton's tyrosine kinase inhibitors; CAR Tcell therapy,
chimeric antigen receptorengineered Tcell therapy.
102
-
DREGER
9. Robinson SP, Boumendil A, Finel H, et al. Longterm outcome anal-
ysis of reducedintensity allogeneic stem cell transplantation in
patients with mantle cell lymphoma: a retrospective study from
the EBMT Lymphoma Working Party. Bone Marrow Transpl.
2018;53(5):617624.
10. Visco C, Tisi MC, Evangelista A, et al. Time to progression
of mantle cell lymphoma after highdose cytarabinebased
regimens defines patients risk for death. Br J Haematol. 2019;
185(5):940944.
11. d'Amore F, Gaulard P, Trümper L, et al. Peripheral Tcell lymphomas:
ESMO clinical practice guidelines for diagnosis, treatment and
followup. Ann Oncol. 2015;26(Suppl 5):v108v115.
12. Schmitz N, Lenz G, Stelljes M. Allogeneic hematopoietic stem
cell transplantation for Tcell lymphomas. Blood. 2018;132(3):
245253.
13. Chihara D, Fanale MA, Miranda RN, et al. The survival outcome of
patients with relapsed/refractory peripheral Tcell lymphomanot
otherwise specified and angioimmunoblastic Tcell lymphoma. Br J
Haematol. 2017;176(5):750758.
14. Schmitz N, Truemper L, Bouabdallah K, et al. A randomized phase 3
trial of auto vs. allo transplantation as part of firstline therapy in
poorrisk peripheral TNHL. Blood. 2020:137. https://doi.org/10.
1182/blood.2020008825
15. Pasquini MC, Hu ZH, Curran K, et al. Realworld evidence of tisa-
genlecleucel for pediatric acute lymphoblastic leukemia and non
Hodgkin lymphoma. Blood Adv. 2020;4(21):54145424.
How to cite this article: Dreger P. Allogeneic stem cell
transplant in nonHodgkin lymphomas: Still an indication?
Hematological Oncology. 2021;39(S1):100103. https://doi.
org/10.1002/hon.2845
DREGER
-
103
Received: 17 February 2021
DOI: 10.1002/hon.2844
SUPPLEMENT ARTICLE
Optimizing CAR T cell therapy in lymphoma
David Qualls |Gilles Salles
Lymphoma Service, Department of Medicine,
Memorial Sloan Kettering Cancer Center, New
York, New York, USA
Correspondence
Gilles Salles, Lymphoma Service, Department
of Medicine, Memorial Sloan Kettering Cancer
Center, New York, New York, USA.
Email: sallesg@mskcc.org
Abstract
Chimeric antigen receptor (CAR) T cell therapy has significantly improved the
outlook for patients with certain types of poorrisk lymphoma. Despite these ad-
vances, a majority of patients undergoing CAR T therapy will suffer progression or
relapse of disease, and toxicity remains a concern. Additionally, the patients and
disease subtypes that are most likely to benefit from CAR T have yet to be fully
defined. Many ongoing trials are exploring novel CAR T approaches to address these
concerns. In this review, we highlight some of the primary strategies and relevant
studies aimed at improving the utility of CAR T therapy in lymphoma.
KEYWORDS
cellular therapy, chimeric antigen receptor T cell, lymphoma
1
|
INTRODUCTION
In 2017, the first chimeric antigen receptor (CAR) T cell therapy was
approved for use in relapsed or refractory aggressive B cell lym-
phoma. The initial CAR T studies demonstrated remarkable efficacy
in a population of patients with otherwise dismal outcomes. In the
4 years since, there has been a marked expansion in the use of CAR T
cell therapy and the development of novel CAR T strategies for the
management of lymphoma.
Despite these encouraging results, significant challenges and
opportunities for improvement in CAR T therapy remain. While the
majority of patients attain responses to treatment, most patients will
eventually develop relapsed or refractory disease. Toxicity also
remains a major limitation to CAR T therapy; cytokine release
syndrome (CRS) and immune effector cellassociated neurotoxicity
syndrome (ICANS) often require hospitalization and present a major
source of cost and morbidity.
With the goal of addressing these shortcomings, over 200 clinical
trials investigating CAR T therapy in lymphoma are underway, with
more strategies in preclinical development (clinicaltrials.gov). This
review aims to highlight some of the most promising strategies for
improving the outcomes of lymphoma patients using CAR T therapy.
Some of the greatest opportunities for advancement include broad-
ening the eligible patient pool and identifying those most likely to
benefit from CAR T, expediting CAR T manufacturing, optimizing
lymphodepletion, improving CAR T design, and coadministering
agents that may improve the efficacy or reduce the toxicity of CAR T
therapy.
2
|
CURRENT PRACTICE
At time of writing, three products, axicabtagene ciloleucel (axicel),
tisagenlecleucel (tisacel), and most recently, lisocabtagene mar-
aleucel (lisocel), are available for commercial use in relapsed or
refractory large B cell lymphoma in various countries. Results with
each agent in their index clinical trials are summarized in Table 1.
Encouraging overall response rates ranging from 52%–82% were
reported, with 12month overall survival of 48%59%.
1–3
These re-
sults compared favorably with historical outcomes in relapsed or
refractory DLBCL, where patients undergoing conventional salvage
therapy had a median overall survival of 6.3 months.
4
All three agents
had characteristic toxicity profiles, with severe (grade 3) CRS in
1%–22% of patients, and severe (grade 3) neurotoxicity in 12%–
28% of patients. Similar efficacy and toxicity outcomes have been
noted in realworld analyses where patients were, on average, older
and with more comorbidities. Trial design and outcomes are reviewed
in greater detail in Table 1, and more detailed analysis of these three
products have recently been published.
5
Most recently, brexucabtagene autoleucel (brexucel) demon-
strated promising results in relapsed or refractory mantle cell
lymphoma (MCL) and is now commercially available.
6
In the ZUMA2
104
-
Hematological Oncology. 2021;39(S1):104112. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
trial, brexucel demonstrated an overall response rate (ORR) of 93%,
and prolonged 12month progressionfree and overall survival at
61% and 83%, respectively. Most patients had CRS and neurotoxicity,
with 15% developing grade 3 CRS and 31% having grade 3 ICANS.
3
|
PATIENT SELECTION
At time of publication, there are no reliable predictive models to
anticipate treatment efficacy or toxicity in CAR T recipients, and this
remains an active area of investigation. The only readily reproducible
determinant to date associated with response is the degree of tumor
burden, as reflected by tumor volume or lactate dehydrogenase
(LDH). Other data suggest that patients with an altered performance
status (Eastern Cooperative Oncology Group (ECOG) >1) or a
heightened inflammatory state (Creactive protein elevation) have a
higher risk of treatment failure.
7,8
Factors classically associated with
prognosis with standard chemoimmunotherapy, such as age, prior
therapy, International Prognostic Index (IPI), and cell of origin, are
not predictive of outcome in CAR T therapy. Other tumor
biomarkers, including immune checkpoint molecules such as PDL1
status, LAG3 expression, and TIM3 have been investigated with
conflicting results in their predictive capacity.
9
A major determinant of response may be the quality and
composition of the patient's autologous T cells that are engineered
into CAR T cells. In one study of chronic lymphocytic leukemia (CLL)
patients, sustained remission was seen in patients with higher
proportions of a T cell subset with CD45RO
CD27
+
CD8
+
immuno-
phenotype, characterized by longlived memory cells in a resting
state, with high capacity for expansion and acquisition of effector
functions on antigen reexposure.
10
CD27
+
PD1
CD8
+
cells with
high expression of IL6R were also predictive of response, providing
potentially intriguing rationale for the interruption of the PD1/PD
L1 axis via CAR T modification or concomitant checkpoint blockade.
T cell polyfunctionality, defined as the capacity of a T cell subset to
deploy multiple immune programs, has also been associated with
both increased overall response and toxicities.
11
These factors may
not only provide prognostic guidance, but also help refine our prac-
tices in preservation and collection of an adequate T cell product. T
cell fitness is compromised by chemotherapy, particularly regimens
incorporating bendamustine, and avoiding such regimens or collect-
ing prior to such treatments may allow for more effective CAR T
products.
12
4
|
EXPANDING CAR T INDICATIONS
Commercially available CAR T products are approved for use in
relapsed or refractory diffuse large B cell lymphoma, highgrade large
B cell lymphoma, transformed follicular lymphoma, and MCL. Intui-
tively these constructs, all targeting the CD19 cell surface antigen,
are likely to have efficacy in other B cell lymphomas, and ongoing
TABLE 1Landmark clinical trials in CAR T cell therapy for lymphoma
ZUMA1 JULIET TRANSCEND ZUMA2
CAR T cell
product
Axicabtagene ciloleucel
(axicel, Yescarta)
Tisagenlecleucel (tisacel,
Kimriyah)
Lisocabtagene maraleucel (lisocel,
JCAR017)
Brexucabtagene autoleucel (brexucel,
Tecartus)
Bridging therapy None 93% 72% 37%
Lymphodepletion Flu/Cy 500/30 3d Flu/Cy 250/25 3d versus
Benda
Flu/Cy 300/30 3d Flu/Cy 500/30 3d
Construct AntiCD19 scFV, CD28,
CD3ζ
AntiCD19 scFV, 41BB,
CD3ζ
AntiCD19 scFV, 41BB, CD3ζAntiCD19 scFV, CD28, CD3ζ
Disease
indications
DLBCL, HGBCL, tFL,
PMBCL
DLBCL, HGBCL, tFL DLBCL, HGBCL, tFL, PMBCL, FL
grade 3B
MCL
Prior lines of
therapy
2222 (prior antiCD20, anthracycline,
BTKi required)
ORR 82% 52% 80% 93%
12 month PFS 44% (34–53) N/A 44.1% (37.3–50.7) 61%
12 month OS 59% (49–68) 48.2% (38.6–57.1) 57.9% (51.3–63.8) 83%
CRS (Grade 3) 93% (13%) 58% (22%) 35% (1%) 91% (15%)
ICANS (Grade
3)
64% (28%) 21% (12%) 19% (13%) 63% (31%)
Note: These studies have resulted in regulatory approval and commercial availability of their respective CAR T products.
Abbreviations: Benda, bendamustine; BTKi, bruton tyrosine kinase inhibitor; CAR, chimeric antigen receptor; CRS, cytokine release syndrome; d, days;
DLBCL, diffuse large Bcell lymphoma; Flu/Cy, fludarabine/cyclophosphamide; HGBCL, high grade Bcell lymphoma; ICANS, immune effector cell
associated neurotoxicity syndrome; MCL, mantle cell lymphoma; N/A, not available; ORR, overall response rate; OS, overall survival; PFS, progression
free survival; PMBCL, primary mediastinal Bcell lymphoma; scFV, single chain variable fragment; tFL, transformed follicular lymphoma.
QUALLS AND SALLES
-
105
studies have demonstrated encouraging preliminary results in other
subtypes, including indolent diseases such as follicular lymphoma,
marginal zone lymphoma, and chronic lymphocytic leukemia.
While indolent lymphomas can often be observed or controlled
with firstline therapies, many patients will ultimately develop
multiply relapsed disease refractory to currently available treat-
ments. Preliminary data (Table 2) from ongoing Phase 2 studies
evaluating axicel, tisacel, and lisocel have demonstrated promising
efficacy, though final results and Food and Drug Administration
(FDA) review are pending.
13–16
Outside the realm of Bcell nonHodgkin lymphoma, CAR T cell
therapy is being explored in Hodgkin lymphoma and T cell lymphoma.
An antiCD30 CAR T cell therapy was recently developed with
encouraging preliminary data for Hodgkin lymphoma.
17
While T cell
lymphoma is an intriguing target for CAR T therapy, there are a
number of unique challenges to this approach. While patients can
tolerate the sustained Bcell aplasia associated with antiCD19
therapy, a similar elimination of the T cell pool would likely cause
severe immunosuppression and lead to infectious complications. As
such, CAR T cells must target cell surface antigens that are dispro-
portionately expressed on the T cell malignancy as compared to the
normal T cell repertoire in order to mitigate “ontarget, offtumor”
toxicity. Targets of interest at this time include CD5, CD7, TRBC1,
CD37, CD38, and B7H3, among others. Another challenge is
CAR T cell fratricide—the possibility of CAR T cells recognizing a
selfantigen, triggering cytotoxic killing of other CAR T cells. One
potential solution is to alter expression of the target antigen on the
CAR T cells themselves.
Alongside histologic indications, the use of CAR T therapy in CNS
lymphoma is gradually expanding. The initial landmark trials for axi
cel and tisacel excluded patients with active CNS disease, given
unclear efficacy and potential for neurotoxicity. Since then, small
studies had shown that commercial CAR T therapy off trial demon-
strated efficacy in secondary CNS lymphoma with no evidence of
excess neurotoxicity.
18,19
The TRANSCEND phase II study of lisocel
included seven patients (3%) with secondary CNS lymphoma, with
three of six evaluable patients achieving complete response.
3
Studies
incorporating patients with primary CNS lymphoma are also
underway.
5
|
INCLUDING CAR T IN EARLIER LINES OF
THERAPY
The commercial use of CAR T therapy is currently approved in the
setting of multiply relapsed/refractory disease. Patients with
aggressive lymphomas who have failed two prior lines of chemo-
immunotherapy are more likely to have complications from their
disease or prior treatment; as a result, they may have more extensive
disease or greater comorbidities at the time of CAR T therapy.
Understandably, there is interest in determining whether CAR T
therapy in an earlier phase of treatment is appropriate. Three
ongoing Phase 3 clinical trials are investigating the three main CAR T
products in the second line setting. In these studies, patients with
relapsed or refractory DLBCL after firstline chemoimmunotherapy
that are transplant eligible are randomized to receive either upfront
CAR T cell therapy or the current standard of care, salvage chemo-
immunotherapy followed by autologous stem cell transplantation.
20
Axicel CAR T therapy is also being investigated in the firstline
setting for those with aggressive B cell lymphoma with initial
adverse characteristics in ZUMA12.
21
In this Phase 2 study, pa-
tients with positive interim PET after two cycles chemo-
immunotherapy underwent leukapheresis followed by
lymphodepletion and CAR T therapy. Interim analysis of 32 patients
revealed an ORR of 85% and CR rate of 74%, which compares
favorably with historical controls.
21
6
|
EXPEDITING THE COLLECTION,
MODIFICATION AND EXPANSION OF CAR T CELL
THERAPY
A significant challenge to the implementation of CAR T therapy is the
delay required for collection, modification, and expansion of CAR T
cells. Median turnaround time from apheresis to infusion in a real
world analysis of commercial therapy was 28 days for axicel and
44 days for tisacel in the United States.
22
Fourteen (9%) of patients
receiving axicel and 3 (4%) of those receiving tisacel died from
progression of disease prior to infusion.
Pointofcare CAR T manufacturing, as opposed to the current
model of centralized CAR T manufacturing, has already shown
promise. A closedsystem manufacturing platform capable of T cell
enrichment, CAR vector transduction, washing and expansion at the
treatment site has already been utilized in phase I clinical trials, with
successful production of effective CAR T cells and turnaround time
from apheresis to infusion of 14 days
23,24
Such systems would allow
for more rapid administration of CAR T product to patients and may
prove more costeffective over time.
Another approach is the use of an allogeneic CAR T product
derived from healthy donors, which would bypass the delays and
costs of autologous CAR T cell collection and manufacturing alto-
gether, while also avoiding the use of potentially dysfunctional
patientderived T cells. Challenges to such an approach include the
risk of graft versus host disease (GVHD) and host rejection of the
allogeneic CAR T cell. The ALPHA study is an ongoing phase I trial
investigating ALLO501, an allogeneic antiCD19 CAR T product.
25
ALLO501 is genetically modified to disrupt the TCR alpha con-
stant gene as prevention against GVHD. CD52 is also disrupted to
allow administration of an antiCD52 monoclonal antibody for
hostspecific lymphodepletion, with the goal of preventing CAR T
rejection. Preliminary results of the first nine evaluable patients
showed encouraging initial response rates with ORR 78%, though
three of seven responders subsequently progressed. While dura-
bility of response remains a question, response rates are encour-
aging and such products may be useful as a bridge to other
therapies, such as allogeneic stem cell transplantation. Alongside
106
-
QUALLS AND SALLES
allogeneic CAR T cells, CARmodified allogeneic natural killer (NK)
cells represent a promising novel therapy, and have shown activity
in nonhodgkin lymphoma (NHL).
26
Potential advantages include
the ability to administer HLAmismatched products, and avoidance
of the CRS and neurotoxicity seen in CAR T therapy, though
questions remain about NK cell persistence and longterm disease
control (Table 3).
7
|
OPTIMIZING LYMPHODEPLETION
Lymphodepletion prior to CAR T infusion has been associated with
improved CAR T cell expansion and antitumor efficacy in hematologic
malignancies.
27
It is thought that lymphodepletion benefits CAR T
cell function via several mechanisms, including the clearance of
endogenous lymphocytes that act as “cytokine sinks” and improved
access to cytokines that facilitate T cell expansion, removal of
immunosuppressive cells such as regulatory T cells and myeloid
derived suppressor cells, and improved function of antigen
presenting cells.
28
While lymphodepletion incorporating fludarabine
and cyclophosphamide (Flu/Cy) is an established component of
CAR T therapy, there is little headtohead data to guide the choice
of agents or dosing.
The role of lymphodepletion in modulating the host immune
profile is gaining increasing interest. In one study, an increase in
monocyte chemoattractant protein1 after lymphodepletion, and
higher interleukin7 (IL7) peak, were associated with improved
progressionfree survival (PFS) in CAR T recipients. Higherintensity
Flu/Cy conditioning was associated with this favorable cytokine
profile, suggesting that modification of the lymphodepletion regimen
to achieve an optimal cytokine response may result in better out-
comes.
29
Another recent preclinical study investigating CAR T ther-
apy in models of lung cancer found that including oxaliplatin in the
lymphodepletion regimen increased intratumor macrophage pro-
duction of Tcell recruiting cytokines and improved CAR T cell tumor
infiltration.
30
Whether a similar strategy of employing immunogenic
chemotherapy would be efficacious in B cell malignancies remains to
be seen.
8
|
BUILDING A BETTER CAR
8.1
|
Improving CAR design
CAR T cells are produced by the transduction, via lentivirus or
retrovirus, of a genetically engineered CAR into an autologous T cell
population. All commercially available products utilize a “second
generation” CAR, which consists of an extracellular antigenbinding
single chain variant domain (scFV), which targets the B cell antigen
CD19; hinge and transmembrane domains, often derived from CD8α
or an extension of CD28; a CD3derived activation domain; and one
or more costimulatory domains, most commonly either CD28 or
41BB. Many variations on this CAR T product are being explored in
preclinical and early clinical trials, exploring modifications in the
number and type of costimulatory domains, the use of alternative
targets or dualtargeted CAR products.
TABLE 2Selected clinical trials investigating expanded indications for CAR T therapies, including indolent lymphoma, CLL, and MCL
ZUMA5 TRANSCEND NHL 001 (MCL) TRANSCEND CLL 004 ELARA
Disease indications FL grade 13A, EN MZL MCL CLL FL grade 13A
Prior lines of therapy 2 (antiCD20, alkylating
agent required)
2 (antiCD20, alkylating
agent, BTKi required)
3 (standardrisk disease) 2
(highrisk disease)
2 (antiCD20, alkylating
agent required)
CAR T cell product Axicel Lisocel Lisocel Tisacel
Median followup,
months
17.5 10.9 (DL1), 3.1 (DL2) 18 6.5
Patients enrolled (treated
per protocol)
151 (151) 42 (32) 23 (22) 98 (52)
CRR (95% CI) 76% 59% 45% 71.1% (56.5–84.0)
ORR (95% CI) 92% (85–97) 84% 82% 84.8% (71.1–93.7)
% Patients with ongoing
response
72% at 12 months 50% at 18 months 84.4% at 6 months
CRS (Grade 3) 82% (7%) 50% (3%) N/A (9%) 48% (0%)
ICANS (Grade 3) 60% (19%) 28% (9%) N/A (22%) 10% (2%)
Note: Of note, outcomes of TRANSCEND NHL 001, TRANSCEND CLL 004, and ELARA are preliminary and based on the most recent available published
results at time of writing.
Abbreviations: BTKi, Bruton tyrosine kinase inhibitor; CAR, chimeric antigen receptor; CLL, chronic lymphocytic leukemia; CRR, complete response rate;
CRS, cytokine release syndrome; DL, dose level; DOR, duration of response; EN MZL, extranodal marginal zone lymphoma; FL, follicular lymphoma;
ICANS, immune effector cellassociated neurotoxicity; ITT, intention to treat; NHL, nonhodgkin lymphoma; ORR, overall response rate; PP, per
protocol.
QUALLS AND SALLES
-
107
8.2
|
Finetuning CAR signaling
After binding target antigen, intracellular domains trigger down-
stream signals leading to T cell activation, proliferation, and anti-
tumor cytotoxicity. The degree of downstream activation appears
to be crucial in determining efficacy. Inadequate signaling leads to
insufficient antitumor activity and poor T cell persistence, whereas
excessive or redundant T cell activation can drive CAR T
exhaustion. Several strategies to optimize CAR signaling are un-
derway investigating various components of the CAR construct
(Figure 1).
Socalled “thirdgeneration” CARs have been developed which
carry both CD28 and 41BB costimulatory domains, with unclear
benefit to date. In one Phase 1 trial of patients with NHL, patients
were simultaneously infused with both secondand thirdgeneration
CAR T cells, with the later CAR T cells exhibited greater expansion
and persistence than second generation CAR T cells, particularly in
patients with lower disease burden.
31
Reducing the activation potential may also improve CAR T cell
persistence and limit exhaustion, allowing for more durable
responses. Modification of the CD3ζactivation domain in order to
modulate the activation potential of CARs has yielded intriguing
TABLE 3Challenges and strategies for improvement of CAR T therapy for lymphoma
Challenge Strategies for improvement Examples
Prolonged time to treatment Pointofcare CAR T manufacturing Closedsystem platforms
Offtheshelf allogeneic CAR T product ALLO501
Autologous T cell
dysfunction
Offtheshelf allogeneic CAR T product ALLO501
Avoiding lymphotoxic chemotherapy prior to T
cell collection
CAR T therapy in earlier lines of treatment earlier T cell collection in
highrisk pts
Improving CART trafficking Combine with cytokine TRUCKs coexpressing IL12, IL15, or IL18
Novel lymphodepletion approaches Higher dose fludarabine/cyclophosphamide
Alternative agents (oxaliplatin)
CD19 loss Novel CAR T targets scFVs specific for CD20, CD22, CD79b, BAFFR, kappa light chain,
ROR1
Dual antigen targeting Bispecific CAR T cells: LV20.19: antiCD19 +antiCD20 AUTO3:
antiCD19 +antiCD22
CAR T exhaustion Blocking immune checkpoint signaling Coadministration of checkpoint inhibitors
AntiPD1 or antiPDL1 blocker secretion
PD1 gene silencing
PD1/CD28 “switch receptors”
Modulating downstream signaling Thirdgeneration CAR T cells
CD3 ITAM modulation (1XX)
Other immunomodulators Ibrutinib
Lenalidomide
Recruiting endogenous
immune cells
Local cytokine production TRUCKS coexpressing IL12, IL15, or IL18
Antigen presenting cell licensing Armored CARs secreting CD40L
Toxicity Antiinflammatory agents Anakinra
Other immunomodulators Ibrutinib
Offswitches Inducible Casp9
EGFR surface expression
CD20 expression
Reversible CAR T inhibition Dasatinib
Changes to CAR T design Hinge/linker changes—CD19BBz(86)
GMCSF gene silencing
Abbreviations: CAR, chimeric antigen receptor; GMCSF, granulocytemacrophage colony stimulating factor; ITAM, immunoreceptor tyrosinebased
activation motif; scFV, single chain variable fragment.
108
-
QUALLS AND SALLES
results in preclinical studies. In one preclinical study, inactivation
of the second and third immunoreceptor tyrosinebased activation
motifs (ITAMs) on CD3ζ, leaving only the most proximal ITAM, led
to increased antitumor activity and T cell persistence.
32
This
“1XX” CAR construct is currently being investigated in an ongoing
clinical trial.
The hinge and transmembrane domains also play a role in CAR
signaling. In one such example, the standard CD19BBz(71) product
utilized in tisacel was modified with additional amino acids, length-
ening the extracellular and intracellular transmembrane domains.
33
The resultant CD19BBz(86) product retained cytolytic capacity
while exhibiting markedly decreased cytokine production as
compared to standard CD19BBz. A Phase I clinical trial of CD19
BBz(86) showed comparable antitumor efficacy with remarkably low
toxicity; no patients had greater than grade I CRS/ICANS, and there
were no significant elevations in serum cytokine levels after infusion.
The underlying mechanism for this functional change remains to be
elucidated, but does reinforce the dynamic role of hinge and trans-
membrane domains, and the potential for less toxic CAR T
constructs.
8.3
|
Alternative and dualtargeted CAR T cells
One of the primary mechanisms of resistance to CD19directed CAR
T therapy is antigen escape via CD19 surface antigen loss (induced by
various molecular alterations in the CD19 gene). To address this,
alternative CARs with scFVs recognizing CD20, CD22, CD79b,
BAFFR, kappa light chain, and ROR1 have been developed and are in
various stages of preclinical and clinical development. Bispecific CAR
T cells targeting two surface antigens simultaneously may prevent
resistance via antigen escape, and clinical studies investigating these
therapies are underway. One such study of LV20.19 CAR T cells, with
bispecific CD19 and CD20 targeting, showed promising efficacy and
safety profile (ORR 82%, CR 64%).
24
Notably, none of the patients
who had progression or relapse had loss of CD19. Similarly encour-
aging results have been reported with AUTO3, a bispecific CAR T
expressing antiCD19 and antiCD22 CARs.
34
8.4
|
Armored CARs, TRUCKs, and other co
modifications
In addition to CAR modifications, T cell comodifications are being
investigated to overcome the immunosuppressive tumor microenvi-
ronment, stimulate intrinsic immune response and in some cases
deliver antitumor drugs in a localized fashion.
35
TRUCKs (T cells
Redirected for Universal Cytokine Killing) are CAR T cells engineered
to secrete cytokines such as IL12, IL15, or IL18. In preclinical
models these modifications have been shown to increase cytotoxicity
and CAR T cell expansion, increase resistance to Treg cellmediated
inhibition, and in some cases recruit endogenous immune cells to the
tumor microenvironment. CAR T cells expressing CD40 ligand
upregulate antigen presentation in B cell malignancies and activate
antigenpresenting cells, and may facilitate a sustained endogenous
immune response against tumor. Other models have been developed
using various strategies to disrupt the PD1/PDL1 immune check-
point axis and disinhibit immune effector cells, including the CAR T
cells themselves.
Comodifications meant to mitigate toxicity have also been
developed. Mutations silencing the production of granulocyte
macrophage colony stimulating factor (GMCSF), which activates
FIGURE 1 Selected CAR T cell constructs currently in use or in development. Figure 1A,B are secondgeneration constructs consistent
with tisacel and axicel, respectively. Third generation (Figure 1C) CAR T cells contain two costimulatory domains. The 1XX CAR construct
(Figure 1D) is genetically modified with activation of the second and third ITAMs, reducing redundant activation and downstream signaling.
Modifications to the hinge and transmembrane domains (Figure 1E) have been demonstrated to alter activation potential and cytokine
production. Dualtargeted CAR T cells have also been developed either expressing a single CAR with two scFV domains targeting two separate
surface antigens (Figure 1F) or expressing two separate CARs with differing targets (Figure 1G). The figure was created with BioRender.com.
CAR, chimeric antigen receptor; ITAM, immunoreceptor tyrosinebased activation motif
QUALLS AND SALLES
-
109
monocytes and macrophages, appear to mitigate CRS while
improving efficacy in preclinical models.
In the event of significant CAR Trelated toxicity, mechanisms for
halting CAR T activity are under investigation. Engineered surface
expression of CD20 or EGFR allows monoclonal antibodymediated
killing of CAR T cells via rituximab or cetuximab, respectively. The
engineered “kill switch” protein iCasp9 can be activated through
administration of a small molecule that facilitates dimerization of
iCasp9 and results in CAR T cell apoptosis. Other approaches include
the use of small molecule therapy for reversible inhibition of CAR T
activity, as recently demonstrated in preclinical models with
dasatinib.
36
9
|
COMBINATION THERAPY
A number of clinical studies are underway evaluating the combina-
tion of established CAR T therapies with a number of potentially
additive or synergistic agents thought to enhance the antitumor ac-
tivity of CAR T cells, reduce T cell exhaustion, or mitigate toxicity.
Phase 1/2 trials investigating CAR T products including checkpoint
inhibitors, immunomodulators, and Bruton tyrosine kinase (BTK) in-
hibitors are ongoing. Other treatments intended to reduce the risk or
severity of CRS and neurotoxicity are also underway, with the anti
IL1R agent anakinra undergoing active clinical investigation.
Disruption of the PD1/PDL1 immune checkpoint through PD1
and PDL1 inhibitors has shown activity in many malignancies,
including Hodgkin lymphoma, though their activity in Bcell NHL is
limited. There is interest in using checkpoint inhibitors alongside CAR
T therapy to reduce CAR T cell exhaustion and facilitate more robust
antitumor responses. One study showed that PD1 with pem-
brolizumab used as “rescue” for patients who progressed after CAR T
therapy showed CAR T reexpansion in 9 of 12 patients, with an ORR
of 27%.
37
While modest, the fact that this therapy generated re-
sponses in patients who had otherwise failed CAR T therapy suggests
a role for checkpoint inhibitors. Concomitant therapy with check-
point inhibitors and CAR T is being investigated in a number of
clinical trials, with preliminary results showing adequate safety pro-
files. Larger studies are needed to determine whether such combi-
nations result in superior outcomes.
Concomitant administration of ibrutinib, a BTK inhibitor, has
been demonstrated in preclinical models to reduce CARTrelated
cytokine production. A recent Phase I study of ibrutinib given
alongside CAR T therapy for ibrutinibrefractory CLL showed an
similar response rates and lower CRS severity, with a trend toward
superior CAR T expansion, when compared to another cohort that
did not receive ibrutinib.
38
Larger prospective clinical trials in CLL
and NHL are ongoing to further evaluate these preliminary findings.
Preclinical data also suggest that immunomodulatory agents
such as lenalidomide may also enhance the antitumor effects of CAR
T cells, and ongoing studies are evaluating the safety and efficacy of
this and other immunomodulatory agents alongside established CAR
T constructs.
10
|
CONCLUSIONS
CAR T therapy has revolutionized the management of patients with
Bcell lymphoma. The applicability of CAR T is expected to expand to
include additional lymphoma subtypes, disease states, and phases of
treatment in the coming years. Challenges remain in overcoming
barriers to CAR T treatment, improving efficacy, and minimizing
toxicity. A number of promising solutions are already in preclinical
and early clinical development, with early data suggesting many safe
and effective CAR T regimens. Determining which of these CAR T
strategies are most effective in improving outcomes will become an
increasingly important question, particularly as they are largely
developed in parallel with few headtohead comparisons. Outside of
the CAR T spectrum, other promising modalities, including bispecific
antibodies, are showing promising results, and determining when to
utilize CAR T over these other strategies presents another challenge.
Going forward, thoughtfully designed clinical trials comparing these
strategies, and further translational studies to understand the bio-
logical underpinnings of this novel technology, will be necessary to
ensure continued improvement in outcomes for patients with
lymphoma.
CONFLICT OF INTEREST
David Qualls has no conflicts of interest to disclose. Gilles Salles has
received in the last 12months financial compensations for partici-
pating to advisory boards or consulting from Abbvie, Beigene, BMS/
Celgene, Debiopharm, Genentech/Roche, Genmab, Incyte, Kite/
Gilead, Milteniy, Morphosys, Novartis, Velosbio
ORCID
Gilles Salles https://orcid.org/0000-0002-9541-8666
REFERENCES
1. Locke FL, Ghobadi A, Jacobson CA, et al. Longterm safety and
activity of axicabtagene ciloleucel in refractory large Bcell lym-
phoma (ZUMA1): a singlearm, multicentre, phase 1–2 trial. Lancet
Oncol 2019;20(1):3142. https://doi.org/10.1016/S14702045(18)
308647.
2. Schuster SJ, Bishop MR, Tam CS, et al. Tisagenlecleucel in adult
relapsed or refractory diffuse large Bcell lymphoma. N Engl J Med.
2019;380(1):4556. https://doi.org/10.1056/nejmoa1804980.
3. Abramson JS, Palomba ML, Gordon LI, et al. Lisocabtagene mar-
aleucel for patients with relapsed or refractory large Bcell lym-
phomas (TRANSCEND NHL 001): a multicentre seamless design
study. Lancet. 2020;396(10254):839852. https://doi.org/10.1016/
S01406736(20)313660.
4. Crump M, Neelapu SS, Farooq U, et al. Outcomes in refractory
diffuse large Bcell lymphoma: results from the international
SCHOLAR1 study. Blood. 2017;130(16):18001808. https://doi.org/
10.1182/blood201703769620.
5. Vitale C, Strati P. CAR Tcell therapy for Bcell nonHodgkin lym-
phoma and chronic lymphocytic leukemia: clinical trials and real
world experiences. Front Oncol. 2020;10:849. https://doi.org/
10.3389/fonc.2020.00849.
6. Wang M, Munoz J, Goy A, et al. KTEX19 CAR Tcell therapy in
relapsed or refractory mantlecell lymphoma. N Engl J Med.
2020;382(14):13311342. https://doi.org/10.1056/nejmoa1914347.
110
-
QUALLS AND SALLES
7. Nastoupil LJ, Jain MD, Feng L, et al. Standardofcare axicabtagene
ciloleucel for relapsed or refractory large Bcell lymphoma: results
from the US lymphoma CAR T consortium. J Clin Oncol.
2020;38(27):31193128. https://doi.org/10.1200/JCO.19.02104.
8. Vercellino L, Di Blasi R, Kanoun S, et al. Predictive factors of early
progression after CAR Tcell therapy in relapsed/refractory diffuse
large Bcell lymphoma. Blood Adv. 2020;4(22):56075615. https://
doi.org/10.1182/bloodadvances.2020003001.
9. Du M, Hari P, Hu Y, Mei H. Biomarkers in individualized manage-
ment of chimeric antigen receptor T cell therapy. Biomark Res.
2020;8(1):113. https://doi.org/10.1186/s40364020001908.
10. Fraietta JA, Lacey SF, Orlando EJ, et al. Determinants of response
and resistance to CD19 chimeric antigen receptor (CAR) T cell
therapy of chronic lymphocytic leukemia. Nat Med. 2018;24(5):
563571. https://doi.org/10.1038/s4159101800101.
11. Rossi J, Paczkowski P, Shen YW, et al. Preinfusion polyfunctional
antiCD19 chimeric antigen receptor T cells are associated with
clinical outcomes in NHL. Blood. 2018;132(8):804814. https://doi.
org/10.1182/blood201801828343.
12. Duell J, Lukic DS, Karg M, et al. Functionally defective T cells after
chemotherapy of Bcell malignancies can Be activated by the tetra-
valent bispecific CD19/CD3 antibody AFM11. J Immunother. 2019;
42(5):180188. https://doi.org/10.1097/CJI.0000000000000267.
13. Jacobson C, Chavez JC, Sehgal AR, et al. Primary analysis of Zuma5:
a phase 2 study of axicabtagene ciloleucel (AxiCel) in patients with
relapsed/refractory (R/R) indolent nonHodgkin lymphoma (iNHL).
Blood. 2020;136(Suppl 1):4041. https://doi.org/10.1182/BLOOD
2020136834.
14. Palomba ML, Gordon LI, Siddiqi T, et al. Safety and preliminary
efficacy in patients with relapsed/refractory mantle cell lymphoma
receiving lisocabtagene maraleucel in transcend NHL 001.
Blood. 2020;136(Suppl 1):1011. https://doi.org/10.1182/blood
2020136158.
15. Siddiqi T, Soumerai JD, Dorritie KA, et al. Updated followup of
patients with relapsed/refractory chronic lymphocytic leukemia/
small lymphocytic lymphoma treated with lisocabtagene maraleucel
in the phase 1 monotherapy cohort of transcend CLL 004, including
highrisk and ibrutinibtreated patients. Blood. 2020;136(Suppl
1):4041. https://doi.org/10.1182/BLOOD2020140491.
16. Fowler NH, Dickinson M, Dreyling M, et al. Efficacy and safety
of tisagenlecleucel in adult patients with relapsed/refractory
follicular lymphoma: interim analysis of the phase 2 Elara trial.
Blood. 2020;136(Suppl 1):13. https://doi.org/10.1182/blood2020
138983.
17. Ramos CA, Grover NS, Beaven AW, et al. AntiCD30 CART cell
therapy in relapsed and refractory Hodgkin lymphoma. J Clin
Oncol. 2020;38(32):37943804. https://doi.org/10.1200/JCO.20.
01342.
18. Frigault MJ, Maus MV, Dietrich J, et al. Tisagenlecleucel CAR Tcell
therapy in secondary CNS lymphoma. Blood. 2019;134(11):860866.
https://doi.org/10.1182/blood.2019001694.
19. Bennani NN, Maurer MJ, Nastoupil LJ, et al. Experience with axi-
cabtagene ciloleucel (Axicel) in patients with secondary CNS
Involvement: results from the US lymphoma CAR T consortium.
Blood. 2019;134(Suppl 1):763763. https://doi.org/10.1182/blood
2019129097.
20. Locke FL. CAR Tcell therapy hopes to move to second line in diffuse
large Bcell lymphoma | the hematologist | American Society of
Hematology. Hematol. 2020;17(6).
21. Neelapu SS, Dickinson M, Ulrickson ML, et al. Interim analysis of
ZUMA12: a phase 2 study of axicabtagene ciloleucel (AxiCel) as
firstline therapy in patients (Pts) with highrisk large B cell lym-
phoma (LBCL). Blood. 2020;136(Suppl 1):4949. https://doi.org/
10.1182/blood2020134449.
22. Riedell PA, Walling C, Nastoupil LJ, et al. A multicenter retrospective
analysis of outcomes and toxicities with commercial axicabtagene
ciloleucel and tisagenlecleucel for relapsed/refractory aggressive
Bcell lymphomas. Biol Blood Marrow Transpl. 2020;26(3):S41S42.
https://doi.org/10.1016/j.bbmt.2019.12.108.
23. Borchmann P, Jühling A, Gödel P, et al. Phase I trial of MB
CART2019.1, a novel CD20 and CD19 targeting tandem chimeric
antigen receptor, in patients with relapsed or refractory Bcell non
Hodgkin lymphoma. Blood. 2020;136(Supplement 1):4848. https://
doi.org/10.1182/blood2020136067.
24. Shah NN, Johnson BD, Schneider D, et al. Bispecific antiCD20, anti
CD19 CAR T cells for relapsed B cell malignancies: a phase 1 dose
escalation and expansion trial. Nat Med. 2020;26(10):15691575.
https://doi.org/10.1038/s4159102010813.
25. Neelapu SS, Munoz J, Locke FL, et al. Firstinhuman data of ALLO
501 and ALLO647 in relapsed/refractory large Bcell or follicular
lymphoma (R/R LBCL/FL): ALPHA study. J Clin Oncol.
2020;38(15_suppl):80028002. https://doi.org/10.1200/jco.2020.38.
15_suppl.8002.
26. Liu E, Marin D, Banerjee P, et al. Use of CARtransduced natural
killer cells in CD19positive lymphoid tumors. N Engl J Med.
2020;382(6):545553. https://doi.org/10.1056/NEJMoa1910607.
27. Brentjens RJ, Rivière I, Park JH, et al. Safety and persistence of
adoptively transferred autologous CD19targeted T cells in patients
with relapsed or chemotherapy refractory Bcell leukemias. Blood.
2011;118(18):48174828. https://doi.org/10.1182/blood201104
348540.
28. Bechman N, Maher J. Lymphodepletion strategies to potentiate
adoptive Tcell immunotherapy what are we doing; where are we
going? Expert Opin Biol Ther. 2011;18:111. https://doi.org/10.1080/
14712598.2021.1857361.
29. Hirayama AV, Gauthier J, Hay KA, et al. The response to lymphode-
pletion impacts PFS in patients with aggressive nonHodgkin
lymphoma treated with CD19 CAR T cells. Blood. 2019;133(17):1876
1887. https://doi.org/10.1182/blood201811887067.
30. Srivastava S, Furlan SN, JaegerRuckstuhl CA, et al. Immunogenic
chemotherapy enhances recruitment of CART cells to lung tumors
and improves antitumor efficacy when combined with checkpoint
blockade. Canc Cell. December 2020. https://doi.org/10.1016/j.
ccell.2020.11.005.
31. Ramos CA, Rouce R, Robertson CS, et al. Vivo fate and activity of
secondversus thirdgeneration CD19specific CART cells in B cell
nonHodgkin's lymphomas. Mol Ther. 2018;26(12):27272737.
https://doi.org/10.1016/j.ymthe.2018.09.009.In
32. Feucht J, Sun J, Eyquem J, et al. Calibration of CAR activation potential
directs alternative T cell fates and therapeutic potency. Nat
Med. 2019;25(1):8288. https://doi.org/10.1038/s41591018
02905.
33. Ying Z, Huang XF, Xiang X, et al. A safe and potent antiCD19 CAR T
cell therapy. Nat Med. 2019;25(6):947953. https://doi.org/10.1038/
s4159101904217.
34. Osborne W, Marzolini M, Tholouli E, et al. Phase I Alexander study
of AUTO3, the first CD19/22 dual targeting CAR T cell therapy, with
pembrolizumab in patients with relapsed/refractory (r/r) DLBCL.
J Clin Oncol. 2020;38(15_suppl):8001. https://doi.org/10.1200/
jco.2020.38.15_suppl.8001.
35. Rafiq S, Hackett CS, Brentjens RJ. Engineering strategies to over-
come the current roadblocks in CAR T cell therapy. Nat Rev Clin
Oncol. 2020;17(3):147167. https://doi.org/10.1038/s41571019
0297y.
36. Mestermann K, Giavridis T, Weber J, et al. The tyrosine kinase in-
hibitor dasatinib acts as a pharmacologic on/off switch for CAR T
cells. Sci Transl Med. 2019;11(499). eaau5907. https://doi.org/
10.1126/scitranslmed.aau5907.
QUALLS AND SALLES
-
111
37. Chong EA, Svoboda J, Dwivedy Nasta S, et al. Sequential antiCD19
directed chimeric antigen receptor modified Tcell therapy
(CART19) and PD1 blockade with pembrolizumab in patients with
relapsed or refractory Bcell nonHodgkin lymphomas. Blood.
2018;132(Suppl 1):4198. https://doi.org/10.1182/blood201899
119502.
38. Gauthier J, Hirayama AV, Purushe J, et al. Feasibility and efficacy of
CD19targeted CAR T cells with concurrent ibrutinib for CLL after
ibrutinib failure. Blood. 2020;135(19):16501660. https://doi.org/
10.1182/blood.2019002936.
How to cite this article: Qualls D, Salles G. Optimizing CAR T
cell therapy in lymphoma. Hematological Oncology.
2021;39(S1):104112. https://doi.org/10.1002/hon.2844
112
-
QUALLS AND SALLES
Received: 3 March 2021
DOI: 10.1002/hon.2858
SUPPLEMENT ARTICLE
Bispecific antibodies for the treatment of lymphomas:
Promises and challenges
Stephen J. Schuster
1,2
1
Perelman School of Medicine, University of
Pennsylvania, Philadelphia, Pennyslvania, USA
2
Lymphoma Program & Lymphoma
Translational Research, Abramson Cancer
Center, University of Pennsylvania,
Philadelphia, Pennyslvania, USA
Correspondence
Stephen J. Schuster, Perelman Center for
Advanced Medicine, 34th & Civic Center
Blvd., 12 Fl. South Pavilion, Philadelphia, PA
19104, USA.
Email: stephen.schuster@pennmedicine.
upenn.edu
Abstract
The potential of bispecific antibodies to direct antigenspecific T cellmediated
cytotoxicity toward malignant cells bearing a target antigen was recognized over
35 years ago. Generally, this is accomplished by combining a Tcell receptorspecific
monoclonal antibody or monoclonal antibodyderived fragment that is capable of
activating and expanding resting T cells with a second monoclonal antibody or
monoclonal antibody fragment directed against a tumor target antigen. Bispecific
antibodies induce effector T cells that bind to tumor cells independently of their T
cell receptor specificity and without the requirement of MHCmediated antigen
presentation, focusing effector Tcell cytotoxicity on tumor cells bearing the target
antigen. The therapeutic efficacy of this approach for treatment of relapsed or re-
fractory Bcell lymphomas was first demonstrated with blinatumomab, a single
molecule comprised of two linked singlechain variable fragments with binding
specificities for CD19 and CD3. The recent demonstration that chimeric antigen
receptor (CAR) modified T cells can achieve very durable remissions in some pa-
tients with relapsed or refractory Bcell lymphomas, as well as the potential efficacy
of bispecific antibodies in CAR T cell failures, has rekindled interested in bispecific
antibodies as a T cellmediated therapeutic approach. We review the early results of
phase 1 clinical trials of bispecific antibodies targeting CD20 on B cells and engaging
T cells via CD3 in 1:1 or 2:1 CD20:CD3 Fab formats for treatment of relapsed or
refractory Bcell lymphomas.
KEYWORDS
bispecific antibody, blinatumomab, glofitamab, immunotherapy, mosunetuzumab,
odronextumab
The term bispecific antibody (bsAb) refers to an antibody or an
antibodyderived protein construct that has binding specificities for
two different antigens. In humans and most mammals, all naturally
occurring monomeric antibodies are bivalent with respect to number
of antigenbinding sites. These two binding sites are almost always
monospecific (i.e., having the same antigenbinding specificity at each
of an antibody's two antigenbinding regions). In comparison, bispe-
cific antibodies combine two different monospecific antigenbinding
regions, or variable regions, from different antibodies to achieve a
single antibody or antibodyderived molecule with bispecific antigen
binding.
In 1960, nearly simultaneous with Porter's seminal observation
that digestion of rabbit antibody with papain yielded two almost
identical univalent antigenbinding fragments (Fab’) per antibody
molecule, Nisonoff and coworkers observed that pepsin digestion
produced a single bivalent antigenbinding fragment [F(ab’)2] which,
after treatment with a disulfidesplitting reagent, yielded two univa-
lent Fab’ fragments that could be recombined into an F(ab’)2 by
Hematological Oncology. 2021;39(S1):113116. wileyonlinelibrary.com/journal/hon © 2021 John Wiley & Sons Ltd.
-
113
FIGURE 1 Antibody fragments—the building blocks of bispecific antibodies
TABLE 1Structure of selected bispecific antibodies
114
-
SCHUSTER
oxidation of sulfhydryl groups.
1,2
(Figure 1). Immediately recognizing
the significance of their observations, these investigators speculated,
“It should also be of interest to attempt to prepare antibody of mixed
specificity.”
2
Soon afterwards in 1964, proof of this concept was
established in an elegant series of experiments by Fudenberg et al. that
demonstrated the successful construction, starting with monospecific
rabbit antisera, of a single “bivalent hybrid antibody” or more precisely
an F(ab’)2 with bispecific binding to both bovine gamma globulin and
chicken ovalbumin.
3
Thus began the development of bispecific anti-
bodies and their application for diagnostic and therapeutic purposes.
The confluence of subsequent discoveries in basic and tumor
immunology, technical innovations in laboratory immunology, and
development of recombinant DNA technology over the next 50 years
resulted in a vast variety of approaches for creating antibodies or
singlemolecule antibody derivatives of “mixed specificity.” Using
combinations of immunoglobulinderived proteins or genetic mod-
ules designed to modify the specificity, valency, size, flexibility,
pharmacokinetic and pharmacodynamic properties, and to enhance
large scale production, a panoply of bsAb formats has evolved and
continues to evolve. Current bsAb formats include full antibody
sized, antibody fragmentsized, and complex larger than antibody
formats, including bivalent and multivalent formats (reviewed in
Brinkmann and Kontermann
4
).
In 1985, the feasibility of directing T cells to effect antigen
specific cytotoxicity toward cells bearing the target antigen of
choice by using “hybrid antibodies” (in this case, chemically linked
monoclonal antibodies) was demonstrated in vitro.
4–6
At that time, it
was recognized that these antibody “heteroconjugates” could (1)
overcome MHCrestricted antigen presentation to cytotoxic T cells,
(2) eliminate the problem of rejection of transplanted effector cells,
and (3) focus effector Tcell cytotoxicity on cells bearing the target
antigen. This was achieved by combining a Tcell receptor specific
monoclonal antibody capable of activating and expanding resting T
cells, as well as inducing effector T cells, together with a second
monoclonal antibody directed against any chosen target antigen.
In 2008, Bargou et al. reported the first use of a bsAb for treat-
ment of lymphomas, namely, blinatumomab.
7
Using two singlechain
variable fragments (Fv, Figure 1; Table 1) with binding specificities
for CD19 +CD3 linked by a serineglycine peptide and administered
by continuous intravenous infusion because of its small size and rapid
clearance, responses to blinatumomab were reported in patients with
relapsed follicular lymphoma, mantle cell lymphoma, and chronic
lymphocytic leukemia. Along with these objective responses, the first
observations of what are now recognized syndromes of special in-
terest related to Tcell engaging therapies—cytokine release syn-
drome (CRS) and immune effector cellassociated neurotoxicity
syndrome (ICANS)—were reported. Subsequent multicenter phase 1
and phase 2 studies in patients with relapsed or refractory Bcell non
Hodgkin lymphomas of various histologies (N=76) and diffuse large B
cell lymphomas (N=25) were completed (Figure 2).
8–10
These studies
confirmed the clinical activity of blinatumomab with overall response
rates of 64% and 43% for varied Bcell lymphomas and for diffuse large
Bcell lymphoma cohorts, respectively; remarkably, 6 of 25 patients
treated at the effective dose of at least 60 µg/m
2
per day remain in
ongoing remissions beyond 5 years (3 with follicular lymphoma, 2 with
mantle cell lymphoma, and 1 with diffuse large Bcell lymphoma).
8
The
acute toxicity profile of blinatumomab, especially neurologic adverse
events, is considerable but manageable, and no late toxicities have
FIGURE 2 Summary of selected bispecific antibodies: safety and efficacy
SCHUSTER
-
115
been observed. In 2014, blinatumomab was approved by the US FDA
for treatment of Philadelphia chromosomenegative relapsed or re-
fractory Bcell precursor acute lymphoblastic leukemia (ALL); its
approval was expanded in 2018 to include minimal residual disease
positive Bcell precursor ALL.
Although the early results for blinatumomab were promising and
supported further development of the bsAb approach for treatment of
Bcell lymphomas, the emergence of CD19directed chimeric antigen
receptor (CAR) modified Tcell therapies (CAR T) over the last 10
years briefly eclipsed the ongoing development of bsAbs.
11,12
How-
ever, the observations that CAR T cells are capable of achieving very
durable remissions in some patients with relapsed or refractory Bcell
lymphomas, as well as the potential efficacy of bsAb in CAR T cell
failures, supported the T cellmediated approach to therapy, whether
by CAR T or bsAb, and ultimately rekindled interested in bsAbs.
Currently, a variety of bsAbs are under development as therapy
for Bcell lymphomas. The early results of phase 1 bsAb clinical trials
are summarized in Table 1and Figure 2. These bsAbs target CD20
on B cells and engage T cells via CD3 in a 1:1 or 2:1 CD20:CD3 Fab
format. They are full size antibody derivatives and thus have phar-
macokinetic profiles that allow intermittent dosing. In general,
neurotoxicity is significantly less frequent than observed with CD19
directed CAR T or blinatumomab therapies. Overall response rates
range from 60 to 90% with complete response rates from 40 to 60%
in relapsed or refractory indolent and follicular lymphomas.
13–15
In
relapsed or refractory large Bcell lymphomas, overall response rates
range from 37% to 90% with complete response rates from 19% to
55%. Importantly, these bsAbs appear to have activity in patients
failing CD19directed CAR T therapy. However, followup for these
new bsAbs is short and the durability of responses remains to be
established.
While proof of principle for bsAbs has been clearly demon-
strated, optimal clinical use of bsAbs in Bcell lymphomas remains to
be established. As we wait for the early phase clinical trials to com-
plete accrual and to mature, we can expect bsAb combinations to be
explored in the next generation of bsAb clinical trials.
CONFLICT OF INTEREST
Schuster is associated with AlloGene (Ad Board), AstraZeneca (Ad
Board), BeiGene (Ad Board) Celgene/BMS (Ad Board, Consulting,
Steering committee), Genentech/Roche (Ad Board, Consulting,
Steering committee, Research Grant), Incyte (Ad Board), Janssen (Ad
Board), Legend Biotech (Ad Board), Loxo Oncology (Ad Board,
Consulting), Novartis (Ad Board, Consulting, Steering committee,
Research Grant), and Regeneron (Ad Board).
DATA AV1AILABILITY STATEMENT
Data sharing is not applicable to this article as no new data were
created or analyzed in this study.
REFERENCES
1. Porter RR. Separation and isolation of fractions of rabbit gamma
globulin containing the antibody and antigenic combining sites. Na-
ture. 1958;182(4636):670671.
2. Nisonoff A, Wissler FC, Lipman LN. Properties of the major
component of a peptic digest of rabbit antibody. Science. 1960;
132(3441):17701771.
3. Fudenberg HH, Drews G, Nisonoff A. Serologic demonstration of
dual specificity of rabbit bivalent hybrid antibody. J Exp Med. 1964;
119(1):151166.
4. Brinkmann U, Kontermann RE. The making of bispecific antibodies.
mAbs. 2017;9(2):182212.
5. Staerz UD, Kanagawa O, Bevan MJ. Hybrid antibodies can target
sites for attack by T cells. Nature. 1985;314(6012):628631.
6. Perez P, Hoffman RW, Shaw S, Bluestone JA, Segal DM. Specific
targeting of cytotoxic T cells by antiT3 linked to antitarget cell
antibody. Nature. 1985;316:354356.
7. Bargou R, Leo E, Zugmaier G, et al. Tumor regression in cancer pa-
tients by very low doses of a T cellengaging antibody. Science.
2008;321(5891):974977.
8. Dufner V, Sayehli CM, Chatterjee M, et al. Longterm outcome of
patients with relapsed/refractory Bcell nonHodgkin lymphoma
treated with blinatumomab. Blood Adv. 2019;3(16):24912498.
9. Goebeler ME, Knop S, Viardot A, et al. Bispecific Tcell engager
(BiTE) antibody construct blinatumomab for the treatment of pa-
tients with relapsed/refractory nonHodgkin lymphoma: final results
from a phase I study. J Clin Oncol. 2016;34(10):11041111.
10. Viardot A, Goebeler ME, Hess G, et al. Phase 2 study of the bis-
pecific Tcell engager (BiTE
®
) antibody blinatumomab in relapsed/
refractory diffuse large Bcell lymphoma. Blood. 2016;127(11):
14101416.
11. Cappell KM, Sherry RM, Yang JC, et al. Longterm followup of anti
CD19 chimeric antigen receptor Tcell therapy. J Clin Oncol.
2020;38(32):38053815.
12. Chong EA, Ruella M, Schuster SJ. Fiveyear outcomes for refractory
Bcell lymphomas with CAR Tcell therapy. N Engl J Med.
2021;384(7):673674.
13. Schuster SJ, Bartlett NL, Assouline S, et al. Mosunetuzumab induces
complete remissions in poor prognosis nonHodgkin lymphoma pa-
tients, including those who are resistant to or relapsing after
chimeric antigen receptor Tcell (CART) therapies, and is active in
treatment through multiple lines. Blood. 2019;134(Suppl 1). ASH
Plenary abstract 6.
14. Hutchings M, CarloStella C, Bachy E, et al. Glofitamab stepup dosing
induces high response rates in patients with hardtotreat refractory
or relapsed nonHodgkin lymphoma. Blood. 2020;136(Supplement
1):4648.
15. Bannerji R, Allan JN, Arnason JE, et al. Odronextamab (REGN1979),
a human CD20 x CD3 bispecific antibody, induces durable, complete
responses in patients with highly refractory Bcell nonHodgkin
lymphoma, including patients refractory to CAR T therapy. Blood.
2020;136(Suppl 1):4243.
How to cite this article: Schuster SJ. Bispecific antibodies for
the treatment of lymphomas: Promises and challenges.
Hematological Oncology. 2021;39(S1):113116. https://doi.
org/10.1002/hon.2858
116
-
SCHUSTER